Symbolic Expert System

In artificial intelligence, symbolic artificial intelligence (likewise called classical expert system or logic-based synthetic intelligence) [1] [2] is the term for the collection of all approaches in synthetic intelligence research study that are based on high-level symbolic (human-readable) representations of issues, reasoning and search. [3] Symbolic AI used tools such as reasoning programs, production rules, semantic nets and frames, and it established applications such as knowledge-based systems (in particular, expert systems), symbolic mathematics, automated theorem provers, ontologies, the semantic web, and automated preparation and scheduling systems. The Symbolic AI paradigm led to influential concepts in search, symbolic programs languages, representatives, multi-agent systems, the semantic web, and the strengths and constraints of formal understanding and reasoning systems.

Symbolic AI was the of AI research study from the mid-1950s until the mid-1990s. [4] Researchers in the 1960s and the 1970s were persuaded that symbolic methods would eventually succeed in developing a machine with artificial basic intelligence and considered this the supreme goal of their field. [citation needed] An early boom, with early successes such as the Logic Theorist and Samuel’s Checkers Playing Program, led to impractical expectations and promises and was followed by the first AI Winter as funding dried up. [5] [6] A second boom (1969-1986) occurred with the rise of professional systems, their pledge of catching corporate knowledge, and an enthusiastic business accept. [7] [8] That boom, and some early successes, e.g., with XCON at DEC, was followed again by later on frustration. [8] Problems with problems in understanding acquisition, keeping large understanding bases, and brittleness in dealing with out-of-domain problems occurred. Another, 2nd, AI Winter (1988-2011) followed. [9] Subsequently, AI scientists focused on addressing hidden issues in dealing with uncertainty and in understanding acquisition. [10] Uncertainty was attended to with official techniques such as covert Markov designs, Bayesian reasoning, and statistical relational knowing. [11] [12] Symbolic maker finding out dealt with the knowledge acquisition issue with contributions consisting of Version Space, Valiant’s PAC learning, Quinlan’s ID3 decision-tree knowing, case-based knowing, and inductive reasoning shows to discover relations. [13]

Neural networks, a subsymbolic technique, had actually been pursued from early days and reemerged highly in 2012. Early examples are Rosenblatt’s perceptron learning work, the backpropagation work of Rumelhart, Hinton and Williams, [14] and work in convolutional neural networks by LeCun et al. in 1989. [15] However, neural networks were not deemed effective until about 2012: “Until Big Data became prevalent, the general agreement in the Al community was that the so-called neural-network approach was helpless. Systems just didn’t work that well, compared to other approaches. … A revolution was available in 2012, when a variety of individuals, consisting of a team of researchers working with Hinton, exercised a way to utilize the power of GPUs to immensely increase the power of neural networks.” [16] Over the next numerous years, deep learning had amazing success in dealing with vision, speech recognition, speech synthesis, image generation, and device translation. However, since 2020, as fundamental difficulties with bias, explanation, comprehensibility, and effectiveness ended up being more apparent with deep learning approaches; an increasing number of AI researchers have required combining the finest of both the symbolic and neural network approaches [17] [18] and resolving areas that both techniques have difficulty with, such as common-sense reasoning. [16]

A brief history of symbolic AI to the present day follows below. Time durations and titles are drawn from Henry Kautz’s 2020 AAAI Robert S. Engelmore Memorial Lecture [19] and the longer Wikipedia short article on the History of AI, with dates and titles varying somewhat for increased clarity.

The very first AI summer: unreasonable spirit, 1948-1966

Success at early attempts in AI happened in 3 main locations: synthetic neural networks, understanding representation, and heuristic search, adding to high expectations. This section summarizes Kautz’s reprise of early AI history.

Approaches inspired by human or animal cognition or habits

Cybernetic approaches tried to replicate the feedback loops in between animals and their environments. A robotic turtle, with sensors, motors for driving and guiding, and seven vacuum tubes for control, based upon a preprogrammed neural internet, was built as early as 1948. This work can be seen as an early precursor to later work in neural networks, reinforcement knowing, and positioned robotics. [20]

A crucial early symbolic AI program was the Logic theorist, composed by Allen Newell, Herbert Simon and Cliff Shaw in 1955-56, as it was able to show 38 primary theorems from Whitehead and Russell’s Principia Mathematica. Newell, Simon, and Shaw later generalized this work to produce a domain-independent problem solver, GPS (General Problem Solver). GPS resolved problems represented with official operators by means of state-space search utilizing means-ends analysis. [21]

During the 1960s, symbolic methods attained excellent success at replicating smart behavior in structured environments such as game-playing, symbolic mathematics, and theorem-proving. AI research study was focused in 4 institutions in the 1960s: Carnegie Mellon University, Stanford, MIT and (later on) University of Edinburgh. Every one established its own design of research study. Earlier approaches based on cybernetics or synthetic neural networks were deserted or pushed into the background.

Herbert Simon and Allen Newell studied human problem-solving abilities and tried to formalize them, and their work laid the foundations of the field of expert system, along with cognitive science, operations research study and management science. Their research team used the outcomes of psychological experiments to establish programs that simulated the methods that people used to resolve issues. [22] [23] This tradition, centered at Carnegie Mellon University would eventually culminate in the development of the Soar architecture in the center 1980s. [24] [25]

Heuristic search

In addition to the highly specialized domain-specific sort of knowledge that we will see later on utilized in professional systems, early symbolic AI scientists discovered another more basic application of understanding. These were called heuristics, general rules that assist a search in appealing instructions: “How can non-enumerative search be practical when the underlying problem is significantly tough? The approach promoted by Simon and Newell is to employ heuristics: quick algorithms that might stop working on some inputs or output suboptimal services.” [26] Another essential advance was to find a method to apply these heuristics that guarantees a solution will be discovered, if there is one, not holding up against the periodic fallibility of heuristics: “The A * algorithm offered a general frame for complete and optimum heuristically guided search. A * is utilized as a subroutine within virtually every AI algorithm today however is still no magic bullet; its guarantee of completeness is purchased the expense of worst-case rapid time. [26]

Early work on knowledge representation and reasoning

Early work covered both applications of official reasoning stressing first-order reasoning, in addition to attempts to handle sensible reasoning in a less formal way.

Modeling official reasoning with logic: the “neats”

Unlike Simon and Newell, John McCarthy felt that machines did not need to imitate the specific systems of human thought, however could rather search for the essence of abstract reasoning and problem-solving with logic, [27] despite whether individuals used the same algorithms. [a] His lab at Stanford (SAIL) focused on utilizing official reasoning to solve a wide range of issues, consisting of understanding representation, preparation and knowing. [31] Logic was likewise the focus of the work at the University of Edinburgh and in other places in Europe which led to the development of the programs language Prolog and the science of reasoning programs. [32] [33]

Modeling implicit common-sense understanding with frames and scripts: the “scruffies”

Researchers at MIT (such as Marvin Minsky and Seymour Papert) [34] [35] [6] found that fixing tough problems in vision and natural language processing needed advertisement hoc solutions-they argued that no easy and basic principle (like logic) would capture all the aspects of intelligent habits. Roger Schank explained their “anti-logic” techniques as “shabby” (rather than the “cool” paradigms at CMU and Stanford). [36] [37] Commonsense knowledge bases (such as Doug Lenat’s Cyc) are an example of “scruffy” AI, since they need to be developed by hand, one complicated concept at a time. [38] [39] [40]

The very first AI winter: crushed dreams, 1967-1977

The very first AI winter season was a shock:

During the first AI summer season, many individuals thought that maker intelligence could be attained in simply a couple of years. The Defense Advance Research Projects Agency (DARPA) introduced programs to support AI research study to use AI to fix issues of national security; in particular, to automate the translation of Russian to English for intelligence operations and to develop self-governing tanks for the battleground. Researchers had started to understand that achieving AI was going to be much harder than was supposed a years earlier, however a mix of hubris and disingenuousness led many university and think-tank scientists to accept financing with promises of deliverables that they must have understood they could not meet. By the mid-1960s neither helpful natural language translation systems nor autonomous tanks had been produced, and a remarkable reaction set in. New DARPA management canceled existing AI financing programs.

Outside of the United States, the most fertile ground for AI research was the UK. The AI winter season in the UK was spurred on not so much by dissatisfied military leaders as by competing academics who viewed AI researchers as charlatans and a drain on research study funding. A teacher of applied mathematics, Sir James Lighthill, was commissioned by Parliament to assess the state of AI research in the nation. The report specified that all of the problems being worked on in AI would be much better dealt with by researchers from other disciplines-such as used mathematics. The report also declared that AI successes on toy problems might never scale to real-world applications due to combinatorial surge. [41]

The 2nd AI summertime: knowledge is power, 1978-1987

Knowledge-based systems

As constraints with weak, domain-independent approaches ended up being increasingly more apparent, [42] researchers from all 3 traditions began to build understanding into AI applications. [43] [7] The understanding revolution was driven by the realization that understanding underlies high-performance, domain-specific AI applications.

Edward Feigenbaum stated:

– “In the knowledge lies the power.” [44]
to explain that high efficiency in a specific domain requires both basic and highly domain-specific understanding. Ed Feigenbaum and Doug Lenat called this The Knowledge Principle:

( 1) The Knowledge Principle: if a program is to carry out a complicated task well, it needs to understand a good deal about the world in which it operates.
( 2) A possible extension of that concept, called the Breadth Hypothesis: there are two extra abilities essential for intelligent behavior in unforeseen circumstances: falling back on progressively basic knowledge, and analogizing to specific however remote understanding. [45]

Success with specialist systems

This “understanding revolution” resulted in the advancement and implementation of specialist systems (presented by Edward Feigenbaum), the first commercially effective type of AI software. [46] [47] [48]

Key expert systems were:

DENDRAL, which found the structure of organic molecules from their chemical formula and mass spectrometer readings.
MYCIN, which identified bacteremia – and suggested further lab tests, when essential – by analyzing lab outcomes, patient history, and medical professional observations. “With about 450 guidelines, MYCIN had the ability to perform in addition to some specialists, and substantially much better than junior doctors.” [49] INTERNIST and CADUCEUS which dealt with internal medication medical diagnosis. Internist tried to record the competence of the chairman of internal medication at the University of Pittsburgh School of Medicine while CADUCEUS might ultimately identify approximately 1000 various diseases.
– GUIDON, which showed how a knowledge base constructed for expert problem solving could be repurposed for mentor. [50] XCON, to set up VAX computers, a then tiresome process that might take up to 90 days. XCON reduced the time to about 90 minutes. [9]
DENDRAL is thought about the very first expert system that relied on knowledge-intensive problem-solving. It is described listed below, by Ed Feigenbaum, from a Communications of the ACM interview, Interview with Ed Feigenbaum:

Among individuals at Stanford interested in computer-based designs of mind was Joshua Lederberg, the 1958 Nobel Prize winner in genetics. When I informed him I desired an induction “sandbox”, he stated, “I have simply the one for you.” His laboratory was doing mass spectrometry of amino acids. The concern was: how do you go from taking a look at the spectrum of an amino acid to the chemical structure of the amino acid? That’s how we began the DENDRAL Project: I was excellent at heuristic search approaches, and he had an algorithm that was great at generating the chemical issue space.

We did not have a grandiose vision. We worked bottom up. Our chemist was Carl Djerassi, developer of the chemical behind the birth control tablet, and also among the world’s most respected mass spectrometrists. Carl and his postdocs were world-class specialists in mass spectrometry. We started to add to their understanding, creating knowledge of engineering as we went along. These experiments amounted to titrating DENDRAL more and more understanding. The more you did that, the smarter the program became. We had excellent outcomes.

The generalization was: in the knowledge lies the power. That was the big concept. In my profession that is the big, “Ah ha!,” and it wasn’t the way AI was being done formerly. Sounds easy, however it’s probably AI’s most powerful generalization. [51]

The other specialist systems mentioned above followed DENDRAL. MYCIN exemplifies the traditional professional system architecture of a knowledge-base of rules combined to a symbolic reasoning system, consisting of making use of certainty elements to deal with uncertainty. GUIDON demonstrates how a specific knowledge base can be repurposed for a second application, tutoring, and is an example of a smart tutoring system, a specific type of knowledge-based application. Clancey revealed that it was not sufficient just to use MYCIN’s rules for instruction, but that he likewise needed to add guidelines for discussion management and trainee modeling. [50] XCON is considerable due to the fact that of the millions of dollars it conserved DEC, which triggered the expert system boom where most all major corporations in the US had skilled systems groups, to catch corporate knowledge, maintain it, and automate it:

By 1988, DEC’s AI group had 40 expert systems released, with more on the way. DuPont had 100 in usage and 500 in advancement. Nearly every significant U.S. corporation had its own Al group and was either utilizing or examining professional systems. [49]

Chess expert understanding was encoded in Deep Blue. In 1996, this enabled IBM’s Deep Blue, with the assistance of symbolic AI, to win in a video game of chess versus the world champion at that time, Garry Kasparov. [52]

Architecture of knowledge-based and skilled systems

A crucial component of the system architecture for all expert systems is the knowledge base, which stores facts and rules for analytical. [53] The simplest technique for a skilled system understanding base is just a collection or network of production guidelines. Production guidelines link signs in a relationship comparable to an If-Then declaration. The professional system processes the rules to make deductions and to determine what extra information it needs, i.e. what questions to ask, utilizing human-readable signs. For example, OPS5, CLIPS and their successors Jess and Drools operate in this style.

Expert systems can operate in either a forward chaining – from proof to conclusions – or backward chaining – from goals to required data and requirements – way. More innovative knowledge-based systems, such as Soar can likewise carry out meta-level reasoning, that is reasoning about their own thinking in regards to deciding how to resolve problems and keeping track of the success of analytical strategies.

Blackboard systems are a 2nd type of knowledge-based or expert system architecture. They design a neighborhood of professionals incrementally contributing, where they can, to fix a problem. The problem is represented in numerous levels of abstraction or alternate views. The professionals (knowledge sources) volunteer their services whenever they acknowledge they can contribute. Potential problem-solving actions are represented on a program that is upgraded as the problem situation changes. A controller chooses how useful each contribution is, and who ought to make the next analytical action. One example, the BB1 blackboard architecture [54] was initially inspired by research studies of how people prepare to carry out multiple tasks in a trip. [55] An innovation of BB1 was to use the very same blackboard model to solving its control problem, i.e., its controller carried out meta-level reasoning with understanding sources that kept an eye on how well a plan or the problem-solving was proceeding and might change from one strategy to another as conditions – such as goals or times – changed. BB1 has actually been used in several domains: building website planning, intelligent tutoring systems, and real-time client monitoring.

The second AI winter, 1988-1993

At the height of the AI boom, companies such as Symbolics, LMI, and Texas Instruments were offering LISP machines specifically targeted to accelerate the development of AI applications and research study. In addition, several expert system business, such as Teknowledge and Inference Corporation, were offering expert system shells, training, and seeking advice from to corporations.

Unfortunately, the AI boom did not last and Kautz best explains the 2nd AI winter that followed:

Many factors can be offered for the arrival of the second AI winter season. The hardware business failed when much more economical basic Unix workstations from Sun together with great compilers for LISP and Prolog came onto the marketplace. Many business deployments of professional systems were discontinued when they proved too costly to maintain. Medical professional systems never captured on for a number of factors: the difficulty in keeping them as much as date; the difficulty for medical professionals to find out how to utilize an overwelming variety of various expert systems for various medical conditions; and maybe most crucially, the reluctance of medical professionals to trust a computer-made diagnosis over their gut instinct, even for specific domains where the specialist systems might exceed a typical physician. Equity capital cash deserted AI almost over night. The world AI conference IJCAI hosted a massive and extravagant trade convention and thousands of nonacademic guests in 1987 in Vancouver; the primary AI conference the list below year, AAAI 1988 in St. Paul, was a small and strictly scholastic affair. [9]

Adding in more rigorous structures, 1993-2011

Uncertain reasoning

Both statistical methods and extensions to reasoning were attempted.

One analytical method, hidden Markov designs, had currently been promoted in the 1980s for speech recognition work. [11] Subsequently, in 1988, Judea Pearl promoted the use of Bayesian Networks as a sound but effective way of dealing with uncertain reasoning with his publication of the book Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. [56] and Bayesian approaches were used effectively in expert systems. [57] Even later, in the 1990s, analytical relational knowing, a method that integrates likelihood with sensible solutions, permitted possibility to be combined with first-order reasoning, e.g., with either Markov Logic Networks or Probabilistic Soft Logic.

Other, non-probabilistic extensions to first-order logic to assistance were also tried. For example, non-monotonic reasoning could be used with reality upkeep systems. A truth upkeep system tracked presumptions and justifications for all reasonings. It enabled reasonings to be withdrawn when assumptions were found out to be inaccurate or a contradiction was obtained. Explanations could be offered a reasoning by discussing which guidelines were used to develop it and after that continuing through underlying inferences and guidelines all the way back to root presumptions. [58] Lofti Zadeh had actually presented a different sort of extension to deal with the representation of ambiguity. For example, in deciding how “heavy” or “tall” a guy is, there is regularly no clear “yes” or “no” response, and a predicate for heavy or tall would instead return values between 0 and 1. Those worths represented to what degree the predicates were true. His fuzzy reasoning even more offered a method for propagating mixes of these worths through rational solutions. [59]

Artificial intelligence

Symbolic machine discovering techniques were investigated to resolve the understanding acquisition traffic jam. Among the earliest is Meta-DENDRAL. Meta-DENDRAL used a generate-and-test strategy to generate plausible rule hypotheses to check versus spectra. Domain and task understanding minimized the variety of prospects checked to a workable size. Feigenbaum described Meta-DENDRAL as

… the culmination of my imagine the early to mid-1960s involving theory development. The conception was that you had an issue solver like DENDRAL that took some inputs and produced an output. In doing so, it used layers of knowledge to steer and prune the search. That understanding acted due to the fact that we talked to people. But how did the people get the understanding? By taking a look at countless spectra. So we wanted a program that would look at countless spectra and infer the understanding of mass spectrometry that DENDRAL might utilize to fix specific hypothesis formation issues. We did it. We were even able to release new knowledge of mass spectrometry in the Journal of the American Chemical Society, offering credit just in a footnote that a program, Meta-DENDRAL, actually did it. We were able to do something that had been a dream: to have a computer system program developed a new and publishable piece of science. [51]

In contrast to the knowledge-intensive method of Meta-DENDRAL, Ross Quinlan created a domain-independent method to analytical category, choice tree learning, starting initially with ID3 [60] and then later on extending its abilities to C4.5. [61] The decision trees created are glass box, interpretable classifiers, with human-interpretable category guidelines.

Advances were made in comprehending artificial intelligence theory, too. Tom Mitchell introduced variation area learning which describes learning as an explore an area of hypotheses, with upper, more basic, and lower, more particular, borders including all viable hypotheses constant with the examples seen up until now. [62] More formally, Valiant introduced Probably Approximately Correct Learning (PAC Learning), a structure for the mathematical analysis of artificial intelligence. [63]

Symbolic machine finding out incorporated more than learning by example. E.g., John Anderson supplied a cognitive design of human learning where skill practice results in a collection of rules from a declarative format to a procedural format with his ACT-R cognitive architecture. For instance, a student might learn to use “Supplementary angles are two angles whose steps sum 180 degrees” as several various procedural rules. E.g., one guideline might say that if X and Y are supplementary and you understand X, then Y will be 180 – X. He called his method “knowledge collection”. ACT-R has actually been used effectively to design elements of human cognition, such as finding out and retention. ACT-R is likewise used in intelligent tutoring systems, called cognitive tutors, to effectively teach geometry, computer system programs, and algebra to school children. [64]

Inductive reasoning shows was another technique to learning that permitted reasoning programs to be synthesized from input-output examples. E.g., Ehud Shapiro’s MIS (Model Inference System) might synthesize Prolog programs from examples. [65] John R. Koza used hereditary algorithms to program synthesis to create hereditary programming, which he used to manufacture LISP programs. Finally, Zohar Manna and Richard Waldinger provided a more basic method to program synthesis that manufactures a practical program in the course of proving its specifications to be proper. [66]

As an alternative to logic, Roger Schank presented case-based reasoning (CBR). The CBR approach outlined in his book, Dynamic Memory, [67] focuses initially on remembering key problem-solving cases for future use and generalizing them where proper. When confronted with a new problem, CBR retrieves the most similar previous case and adapts it to the specifics of the present problem. [68] Another alternative to logic, hereditary algorithms and genetic shows are based upon an evolutionary design of knowing, where sets of rules are encoded into populations, the rules govern the behavior of individuals, and choice of the fittest prunes out sets of unsuitable rules over many generations. [69]

Symbolic artificial intelligence was applied to finding out concepts, rules, heuristics, and problem-solving. Approaches, besides those above, consist of:

1. Learning from guideline or advice-i.e., taking human guideline, positioned as recommendations, and determining how to operationalize it in particular scenarios. For example, in a video game of Hearts, learning exactly how to play a hand to “avoid taking points.” [70] 2. Learning from exemplars-improving efficiency by accepting subject-matter expert (SME) feedback during training. When problem-solving stops working, querying the professional to either learn a brand-new exemplar for problem-solving or to learn a new description regarding exactly why one exemplar is more appropriate than another. For example, the program Protos found out to diagnose ringing in the ears cases by communicating with an audiologist. [71] 3. Learning by analogy-constructing issue solutions based upon similar issues seen in the past, and after that customizing their services to fit a new circumstance or domain. [72] [73] 4. Apprentice knowing systems-learning unique services to problems by observing human problem-solving. Domain knowledge describes why novel solutions are correct and how the solution can be generalized. LEAP learned how to develop VLSI circuits by observing human designers. [74] 5. Learning by discovery-i.e., developing jobs to bring out experiments and then finding out from the results. Doug Lenat’s Eurisko, for instance, discovered heuristics to beat human players at the Traveller role-playing video game for 2 years in a row. [75] 6. Learning macro-operators-i.e., searching for helpful macro-operators to be found out from sequences of basic analytical actions. Good macro-operators streamline problem-solving by enabling problems to be resolved at a more abstract level. [76]
Deep learning and neuro-symbolic AI 2011-now

With the increase of deep learning, the symbolic AI approach has been compared to deep knowing as complementary “… with parallels having been drawn lot of times by AI scientists between Kahneman’s research on human thinking and choice making – shown in his book Thinking, Fast and Slow – and the so-called “AI systems 1 and 2″, which would in concept be modelled by deep knowing and symbolic reasoning, respectively.” In this view, symbolic thinking is more apt for deliberative thinking, preparation, and explanation while deep knowing is more apt for fast pattern acknowledgment in perceptual applications with noisy data. [17] [18]

Neuro-symbolic AI: integrating neural and symbolic methods

Neuro-symbolic AI efforts to incorporate neural and symbolic architectures in a manner that addresses strengths and weak points of each, in a complementary style, in order to support robust AI efficient in thinking, discovering, and cognitive modeling. As argued by Valiant [77] and many others, [78] the reliable building of rich computational cognitive models requires the mix of sound symbolic thinking and efficient (machine) learning designs. Gary Marcus, likewise, argues that: “We can not construct rich cognitive designs in an adequate, automated method without the triune of hybrid architecture, rich anticipation, and sophisticated methods for thinking.”, [79] and in particular: “To develop a robust, knowledge-driven approach to AI we need to have the machinery of symbol-manipulation in our toolkit. Excessive of useful knowledge is abstract to make do without tools that represent and manipulate abstraction, and to date, the only equipment that we know of that can control such abstract understanding dependably is the apparatus of sign adjustment. ” [80]

Henry Kautz, [19] Francesca Rossi, [81] and Bart Selman [82] have also argued for a synthesis. Their arguments are based on a requirement to address the 2 kinds of thinking gone over in Daniel Kahneman’s book, Thinking, Fast and Slow. Kahneman describes human thinking as having two elements, System 1 and System 2. System 1 is fast, automated, user-friendly and unconscious. System 2 is slower, step-by-step, and specific. System 1 is the kind utilized for pattern recognition while System 2 is far much better matched for preparation, reduction, and deliberative thinking. In this view, deep knowing best designs the very first kind of thinking while symbolic thinking finest designs the second kind and both are needed.

Garcez and Lamb explain research in this area as being ongoing for a minimum of the past twenty years, [83] dating from their 2002 book on neurosymbolic knowing systems. [84] A series of workshops on neuro-symbolic reasoning has actually been held every year considering that 2005, see http://www.neural-symbolic.org/ for information.

In their 2015 paper, Neural-Symbolic Learning and Reasoning: Contributions and Challenges, Garcez et al. argue that:

The combination of the symbolic and connectionist paradigms of AI has been pursued by a reasonably small research neighborhood over the last twenty years and has actually yielded several considerable outcomes. Over the last decade, neural symbolic systems have actually been revealed efficient in overcoming the so-called propositional fixation of neural networks, as McCarthy (1988) put it in action to Smolensky (1988 ); see also (Hinton, 1990). Neural networks were revealed efficient in representing modal and temporal logics (d’Avila Garcez and Lamb, 2006) and fragments of first-order reasoning (Bader, Hitzler, Hölldobler, 2008; d’Avila Garcez, Lamb, Gabbay, 2009). Further, neural-symbolic systems have been used to a number of problems in the areas of bioinformatics, control engineering, software verification and adjustment, visual intelligence, ontology knowing, and video game. [78]

Approaches for combination are varied. Henry Kautz’s taxonomy of neuro-symbolic architectures, along with some examples, follows:

– Symbolic Neural symbolic-is the existing method of many neural models in natural language processing, where words or subword tokens are both the supreme input and output of large language designs. Examples consist of BERT, RoBERTa, and GPT-3.
– Symbolic [Neural] -is exemplified by AlphaGo, where symbolic strategies are utilized to call neural strategies. In this case the symbolic approach is Monte Carlo tree search and the neural methods find out how to evaluate game positions.
– Neural|Symbolic-uses a neural architecture to translate affective information as signs and relationships that are then reasoned about symbolically.
– Neural: Symbolic → Neural-relies on symbolic thinking to produce or label training information that is subsequently discovered by a deep knowing model, e.g., to train a neural model for symbolic computation by utilizing a Macsyma-like symbolic mathematics system to create or label examples.
– Neural _ Symbolic -utilizes a neural internet that is created from symbolic guidelines. An example is the Neural Theorem Prover, [85] which constructs a neural network from an AND-OR proof tree produced from knowledge base rules and terms. Logic Tensor Networks [86] also fall into this classification.
– Neural [Symbolic] -enables a neural design to straight call a symbolic reasoning engine, e.g., to perform an action or evaluate a state.

Many key research study questions remain, such as:

– What is the very best method to integrate neural and symbolic architectures? [87]- How should symbolic structures be represented within neural networks and drawn out from them?
– How should common-sense understanding be learned and reasoned about?
– How can abstract knowledge that is hard to encode realistically be handled?

Techniques and contributions

This area provides a summary of techniques and contributions in an overall context leading to many other, more in-depth posts in Wikipedia. Sections on Machine Learning and Uncertain Reasoning are covered previously in the history area.

AI programs languages

The essential AI programming language in the US throughout the last symbolic AI boom duration was LISP. LISP is the second oldest programming language after FORTRAN and was produced in 1958 by John McCarthy. LISP supplied the first read-eval-print loop to support quick program advancement. Compiled functions might be easily blended with interpreted functions. Program tracing, stepping, and breakpoints were likewise offered, along with the capability to alter values or functions and continue from breakpoints or mistakes. It had the very first self-hosting compiler, suggesting that the compiler itself was initially composed in LISP and after that ran interpretively to assemble the compiler code.

Other crucial developments originated by LISP that have spread to other shows languages include:

Garbage collection
Dynamic typing
Higher-order functions
Recursion
Conditionals

Programs were themselves data structures that other programs could run on, enabling the easy definition of higher-level languages.

In contrast to the US, in Europe the key AI programs language during that exact same period was Prolog. Prolog offered an integrated shop of truths and provisions that might be queried by a read-eval-print loop. The shop might act as an understanding base and the provisions could act as guidelines or a limited form of reasoning. As a subset of first-order reasoning Prolog was based upon Horn stipulations with a closed-world assumption-any truths not understood were considered false-and a special name presumption for primitive terms-e.g., the identifier barack_obama was thought about to refer to exactly one item. Backtracking and unification are built-in to Prolog.

Alain Colmerauer and Philippe Roussel are credited as the inventors of Prolog. Prolog is a form of reasoning programming, which was created by Robert Kowalski. Its history was also influenced by Carl Hewitt’s PLANNER, an assertional database with pattern-directed invocation of approaches. For more detail see the section on the origins of Prolog in the PLANNER post.

Prolog is likewise a kind of declarative programming. The logic clauses that explain programs are straight analyzed to run the programs defined. No specific series of actions is needed, as is the case with vital shows languages.

Japan championed Prolog for its Fifth Generation Project, intending to develop unique hardware for high efficiency. Similarly, LISP makers were developed to run LISP, but as the second AI boom turned to bust these business might not compete with new workstations that could now run LISP or Prolog natively at equivalent speeds. See the history section for more detail.

Smalltalk was another influential AI shows language. For instance, it introduced metaclasses and, along with Flavors and CommonLoops, affected the Common Lisp Object System, or (CLOS), that is now part of Common Lisp, the present basic Lisp dialect. CLOS is a Lisp-based object-oriented system that allows multiple inheritance, in addition to incremental extensions to both classes and metaclasses, hence providing a run-time meta-object protocol. [88]

For other AI shows languages see this list of shows languages for expert system. Currently, Python, a multi-paradigm programming language, is the most popular programs language, partially due to its extensive package library that supports data science, natural language processing, and deep knowing. Python consists of a read-eval-print loop, functional aspects such as higher-order functions, and object-oriented shows that includes metaclasses.

Search

Search occurs in numerous type of issue resolving, including planning, restraint fulfillment, and playing games such as checkers, chess, and go. The very best understood AI-search tree search algorithms are breadth-first search, depth-first search, A *, and Monte Carlo Search. Key search algorithms for Boolean satisfiability are WalkSAT, conflict-driven stipulation learning, and the DPLL algorithm. For adversarial search when playing video games, alpha-beta pruning, branch and bound, and minimax were early contributions.

Knowledge representation and reasoning

Multiple various methods to represent knowledge and then reason with those representations have been investigated. Below is a quick summary of approaches to knowledge representation and automated thinking.

Knowledge representation

Semantic networks, conceptual graphs, frames, and logic are all approaches to modeling understanding such as domain understanding, analytical understanding, and the semantic meaning of language. Ontologies model essential ideas and their relationships in a domain. Example ontologies are YAGO, WordNet, and DOLCE. DOLCE is an example of an upper ontology that can be used for any domain while WordNet is a lexical resource that can also be seen as an ontology. YAGO integrates WordNet as part of its ontology, to line up truths drawn out from Wikipedia with WordNet synsets. The Disease Ontology is an example of a medical ontology presently being utilized.

Description logic is a logic for automated category of ontologies and for spotting inconsistent category information. OWL is a language used to represent ontologies with description reasoning. Protégé is an ontology editor that can read in OWL ontologies and then examine consistency with deductive classifiers such as such as HermiT. [89]

First-order logic is more general than description reasoning. The automated theorem provers discussed listed below can prove theorems in first-order reasoning. Horn stipulation reasoning is more limited than first-order logic and is utilized in reasoning programs languages such as Prolog. Extensions to first-order logic include temporal reasoning, to manage time; epistemic reasoning, to reason about agent knowledge; modal reasoning, to manage possibility and need; and probabilistic reasonings to handle reasoning and possibility together.

Automatic theorem proving

Examples of automated theorem provers for first-order reasoning are:

Prover9.
ACL2.
Vampire.

Prover9 can be used in conjunction with the Mace4 model checker. ACL2 is a theorem prover that can manage proofs by induction and is a descendant of the Boyer-Moore Theorem Prover, also referred to as Nqthm.

Reasoning in knowledge-based systems

Knowledge-based systems have a specific understanding base, generally of guidelines, to improve reusability throughout domains by separating procedural code and domain knowledge. A different inference engine processes rules and includes, deletes, or modifies an understanding store.

Forward chaining inference engines are the most typical, and are seen in CLIPS and OPS5. Backward chaining happens in Prolog, where a more restricted sensible representation is utilized, Horn Clauses. Pattern-matching, particularly marriage, is utilized in Prolog.

A more versatile sort of analytical occurs when thinking about what to do next happens, instead of just selecting one of the readily available actions. This kind of meta-level reasoning is used in Soar and in the BB1 chalkboard architecture.

Cognitive architectures such as ACT-R might have extra capabilities, such as the ability to assemble often utilized understanding into higher-level chunks.

Commonsense reasoning

Marvin Minsky initially proposed frames as a way of interpreting typical visual situations, such as a workplace, and Roger Schank extended this concept to scripts for common regimens, such as eating in restaurants. Cyc has tried to record useful sensible understanding and has “micro-theories” to handle specific kinds of domain-specific reasoning.

Qualitative simulation, such as Benjamin Kuipers’s QSIM, [90] approximates human thinking about naive physics, such as what happens when we heat a liquid in a pot on the range. We expect it to heat and possibly boil over, even though we might not know its temperature, its boiling point, or other details, such as atmospheric pressure.

Similarly, Allen’s temporal period algebra is a simplification of reasoning about time and Region Connection Calculus is a simplification of reasoning about spatial relationships. Both can be solved with restraint solvers.

Constraints and constraint-based reasoning

Constraint solvers carry out a more minimal kind of reasoning than first-order logic. They can simplify sets of spatiotemporal constraints, such as those for RCC or Temporal Algebra, in addition to resolving other sort of puzzle issues, such as Wordle, Sudoku, cryptarithmetic problems, and so on. Constraint logic programs can be utilized to resolve scheduling issues, for instance with restriction dealing with guidelines (CHR).

Automated preparation

The General Problem Solver (GPS) cast planning as analytical utilized means-ends analysis to develop strategies. STRIPS took a different method, seeing planning as theorem proving. Graphplan takes a least-commitment technique to preparation, instead of sequentially choosing actions from a preliminary state, working forwards, or a goal state if working in reverse. Satplan is an approach to planning where a planning issue is minimized to a Boolean satisfiability problem.

Natural language processing

Natural language processing concentrates on dealing with language as data to carry out jobs such as determining subjects without always comprehending the intended significance. Natural language understanding, in contrast, constructs a meaning representation and uses that for more processing, such as addressing concerns.

Parsing, tokenizing, spelling correction, part-of-speech tagging, noun and verb expression chunking are all elements of natural language processing long dealt with by symbolic AI, however since enhanced by deep learning methods. In symbolic AI, discourse representation theory and first-order reasoning have been utilized to represent sentence significances. Latent semantic analysis (LSA) and explicit semantic analysis likewise supplied vector representations of documents. In the latter case, vector components are interpretable as principles called by Wikipedia short articles.

New deep knowing techniques based upon Transformer models have actually now eclipsed these earlier symbolic AI approaches and achieved modern performance in natural language processing. However, Transformer models are opaque and do not yet produce human-interpretable semantic representations for sentences and files. Instead, they produce task-specific vectors where the meaning of the vector components is opaque.

Agents and multi-agent systems

Agents are autonomous systems embedded in an environment they perceive and act upon in some sense. Russell and Norvig’s basic textbook on artificial intelligence is arranged to show agent architectures of increasing elegance. [91] The sophistication of representatives differs from simple reactive agents, to those with a model of the world and automated planning abilities, possibly a BDI representative, i.e., one with beliefs, desires, and intentions – or additionally a reinforcement finding out model discovered over time to pick actions – up to a mix of alternative architectures, such as a neuro-symbolic architecture [87] that includes deep learning for perception. [92]

On the other hand, a multi-agent system consists of multiple representatives that communicate amongst themselves with some inter-agent interaction language such as Knowledge Query and Manipulation Language (KQML). The agents require not all have the very same internal architecture. Advantages of multi-agent systems include the ability to divide work among the agents and to increase fault tolerance when agents are lost. Research issues include how representatives reach consensus, dispersed problem resolving, multi-agent knowing, multi-agent preparation, and dispersed restriction optimization.

Controversies occurred from early in symbolic AI, both within the field-e.g., in between logicists (the pro-logic “neats”) and non-logicists (the anti-logic “scruffies”)- and between those who embraced AI but rejected symbolic approaches-primarily connectionists-and those outside the field. Critiques from outside of the field were mainly from thinkers, on intellectual premises, but also from financing companies, especially throughout the two AI winter seasons.

The Frame Problem: knowledge representation difficulties for first-order logic

Limitations were found in using basic first-order logic to factor about vibrant domains. Problems were discovered both with concerns to specifying the preconditions for an action to succeed and in providing axioms for what did not alter after an action was carried out.

McCarthy and Hayes presented the Frame Problem in 1969 in the paper, “Some Philosophical Problems from the Standpoint of Artificial Intelligence.” [93] A simple example occurs in “showing that one person might enter into conversation with another”, as an axiom asserting “if a person has a telephone he still has it after looking up a number in the telephone directory” would be needed for the deduction to prosper. Similar axioms would be needed for other domain actions to specify what did not change.

A comparable issue, called the Qualification Problem, happens in attempting to enumerate the prerequisites for an action to be successful. A boundless number of pathological conditions can be envisioned, e.g., a banana in a tailpipe might avoid a cars and truck from operating correctly.

McCarthy’s technique to repair the frame issue was circumscription, a sort of non-monotonic logic where reductions might be made from actions that require only define what would alter while not having to clearly specify whatever that would not change. Other non-monotonic logics offered reality maintenance systems that modified beliefs resulting in contradictions.

Other methods of managing more open-ended domains included probabilistic reasoning systems and artificial intelligence to learn brand-new concepts and rules. McCarthy’s Advice Taker can be deemed a motivation here, as it might incorporate new understanding supplied by a human in the type of assertions or guidelines. For instance, speculative symbolic maker learning systems checked out the capability to take high-level natural language suggestions and to translate it into domain-specific actionable rules.

Similar to the issues in dealing with dynamic domains, sensible thinking is also hard to catch in formal reasoning. Examples of common-sense reasoning include implicit thinking about how people believe or general knowledge of daily events, items, and living animals. This sort of understanding is taken for granted and not considered as noteworthy. Common-sense reasoning is an open location of research study and challenging both for symbolic systems (e.g., Cyc has attempted to capture key parts of this knowledge over more than a years) and neural systems (e.g., self-driving cars that do not understand not to drive into cones or not to strike pedestrians strolling a bicycle).

McCarthy saw his Advice Taker as having common-sense, however his definition of sensible was various than the one above. [94] He defined a program as having sound judgment “if it instantly deduces for itself an adequately wide class of immediate effects of anything it is informed and what it currently knows. “

Connectionist AI: philosophical challenges and sociological disputes

Connectionist methods include earlier work on neural networks, [95] such as perceptrons; work in the mid to late 80s, such as Danny Hillis’s Connection Machine and Yann LeCun’s advances in convolutional neural networks; to today’s advanced methods, such as Transformers, GANs, and other operate in deep knowing.

Three philosophical positions [96] have actually been laid out amongst connectionists:

1. Implementationism-where connectionist architectures carry out the abilities for symbolic processing,
2. Radical connectionism-where symbolic processing is turned down absolutely, and connectionist architectures underlie intelligence and are fully sufficient to describe it,
3. Moderate connectionism-where symbolic processing and connectionist architectures are deemed complementary and both are needed for intelligence

Olazaran, in his sociological history of the controversies within the neural network community, described the moderate connectionism view as basically suitable with current research study in neuro-symbolic hybrids:

The third and last position I would like to examine here is what I call the moderate connectionist view, a more diverse view of the current argument in between connectionism and symbolic AI. Among the scientists who has actually elaborated this position most explicitly is Andy Clark, a philosopher from the School of Cognitive and Computing Sciences of the University of Sussex (Brighton, England). Clark protected hybrid (partly symbolic, partly connectionist) systems. He declared that (a minimum of) 2 type of theories are required in order to study and model cognition. On the one hand, for some information-processing jobs (such as pattern recognition) connectionism has advantages over symbolic designs. But on the other hand, for other cognitive processes (such as serial, deductive thinking, and generative symbol control procedures) the symbolic paradigm offers sufficient designs, and not just “approximations” (contrary to what extreme connectionists would declare). [97]

Gary Marcus has declared that the animus in the deep learning community against symbolic methods now may be more sociological than philosophical:

To believe that we can simply abandon symbol-manipulation is to suspend disbelief.

And yet, for the most part, that’s how most present AI proceeds. Hinton and numerous others have actually striven to banish signs completely. The deep learning hope-seemingly grounded not so much in science, but in a sort of historic grudge-is that smart behavior will emerge purely from the confluence of enormous data and deep learning. Where classical computer systems and software application solve tasks by specifying sets of symbol-manipulating rules committed to specific jobs, such as editing a line in a word processor or performing an estimation in a spreadsheet, neural networks normally try to solve jobs by analytical approximation and discovering from examples.

According to Marcus, Geoffrey Hinton and his associates have actually been vehemently “anti-symbolic”:

When deep learning reemerged in 2012, it was with a sort of take-no-prisoners attitude that has identified most of the last decade. By 2015, his hostility toward all things signs had actually fully taken shape. He lectured at an AI workshop at Stanford comparing symbols to aether, one of science’s greatest errors.

…

Since then, his anti-symbolic project has just increased in strength. In 2016, Yann LeCun, Bengio, and Hinton wrote a manifesto for deep learning in among science’s most essential journals, Nature. It closed with a direct attack on sign adjustment, calling not for reconciliation but for straight-out replacement. Later, Hinton informed an event of European Union leaders that investing any further money in symbol-manipulating approaches was “a substantial mistake,” comparing it to buying internal combustion engines in the age of electrical automobiles. [98]

Part of these disagreements may be due to unclear terms:

Turing award winner Judea Pearl provides a review of artificial intelligence which, unfortunately, conflates the terms machine learning and deep knowing. Similarly, when Geoffrey Hinton describes symbolic AI, the undertone of the term tends to be that of specialist systems dispossessed of any capability to learn. Using the terminology needs clarification. Artificial intelligence is not restricted to association rule mining, c.f. the body of work on symbolic ML and relational knowing (the differences to deep learning being the choice of representation, localist rational rather than dispersed, and the non-use of gradient-based learning algorithms). Equally, symbolic AI is not practically production rules composed by hand. An appropriate meaning of AI concerns understanding representation and reasoning, autonomous multi-agent systems, planning and argumentation, in addition to learning. [99]

Situated robotics: the world as a model

Another critique of symbolic AI is the embodied cognition approach:

The embodied cognition method claims that it makes no sense to consider the brain individually: cognition takes location within a body, which is embedded in an environment. We require to study the system as a whole; the brain’s working exploits consistencies in its environment, consisting of the rest of its body. Under the embodied cognition approach, robotics, vision, and other sensing units end up being main, not peripheral. [100]

Rodney Brooks developed behavior-based robotics, one approach to embodied cognition. Nouvelle AI, another name for this technique, is deemed an alternative to both symbolic AI and connectionist AI. His method turned down representations, either symbolic or distributed, as not only unneeded, however as destructive. Instead, he developed the subsumption architecture, a layered architecture for embodied representatives. Each layer accomplishes a different function and needs to function in the real life. For instance, the first robot he describes in Intelligence Without Representation, has three layers. The bottom layer interprets finder sensors to prevent objects. The middle layer causes the robotic to roam around when there are no challenges. The top layer causes the robot to go to more far-off places for additional expedition. Each layer can temporarily inhibit or reduce a lower-level layer. He slammed AI researchers for specifying AI problems for their systems, when: “There is no clean division between perception (abstraction) and thinking in the genuine world.” [101] He called his robots “Creatures” and each layer was “composed of a fixed-topology network of basic finite state makers.” [102] In the Nouvelle AI method, “First, it is extremely crucial to check the Creatures we build in the genuine world; i.e., in the same world that we human beings inhabit. It is disastrous to fall under the temptation of testing them in a streamlined world first, even with the best objectives of later transferring activity to an unsimplified world.” [103] His emphasis on real-world testing was in contrast to “Early operate in AI focused on games, geometrical problems, symbolic algebra, theorem proving, and other official systems” [104] and making use of the blocks world in symbolic AI systems such as SHRDLU.

Current views

Each approach-symbolic, connectionist, and behavior-based-has benefits, however has been slammed by the other techniques. Symbolic AI has actually been slammed as disembodied, accountable to the credentials problem, and poor in dealing with the affective problems where deep learning excels. In turn, connectionist AI has actually been slammed as improperly matched for deliberative step-by-step issue resolving, including understanding, and handling planning. Finally, Nouvelle AI masters reactive and real-world robotics domains however has actually been slammed for troubles in including learning and understanding.

Hybrid AIs including one or more of these techniques are presently considered as the course forward. [19] [81] [82] Russell and Norvig conclude that:

Overall, Dreyfus saw locations where AI did not have complete responses and stated that Al is therefore difficult; we now see numerous of these exact same locations going through continued research and development causing increased capability, not impossibility. [100]

Expert system.
Automated preparation and scheduling
Automated theorem proving
Belief modification
Case-based thinking
Cognitive architecture
Cognitive science
Connectionism
Constraint programming
Deep learning
First-order reasoning
GOFAI
History of synthetic intelligence
Inductive reasoning shows
Knowledge-based systems
Knowledge representation and reasoning
Logic programs
Artificial intelligence
Model monitoring
Model-based reasoning
Multi-agent system
Natural language processing
Neuro-symbolic AI
Ontology
Philosophy of artificial intelligence
Physical sign systems hypothesis
Semantic Web
Sequential pattern mining
Statistical relational knowing
Symbolic mathematics
YAGO ontology
WordNet

Notes

^ McCarthy as soon as said: “This is AI, so we do not care if it’s psychologically real”. [4] McCarthy reiterated his position in 2006 at the AI@50 conference where he stated “Expert system is not, by meaning, simulation of human intelligence”. [28] Pamela McCorduck composes that there are “2 major branches of synthetic intelligence: one focused on producing smart behavior despite how it was accomplished, and the other aimed at modeling smart processes discovered in nature, especially human ones.”, [29] Stuart Russell and Peter Norvig wrote “Aeronautical engineering texts do not define the objective of their field as making ‘makers that fly so exactly like pigeons that they can trick even other pigeons.'” [30] Citations

^ Garnelo, Marta; Shanahan, Murray (October 2019). “Reconciling deep learning with symbolic artificial intelligence: representing objects and relations”. Current Opinion in Behavioral Sciences. 29: 17-23. doi:10.1016/ j.cobeha.2018.12.010. hdl:10044/ 1/67796.
^ Thomason, Richmond (February 27, 2024). “Logic-Based Artificial Intelligence”. In Zalta, Edward N. (ed.). Stanford Encyclopedia of Philosophy.
^ Garnelo, Marta; Shanahan, Murray (2019-10-01). “Reconciling deep learning with symbolic expert system: representing items and relations”. Current Opinion in Behavioral Sciences. 29: 17-23. doi:10.1016/ j.cobeha.2018.12.010. hdl:10044/ 1/67796. S2CID 72336067.
^ a b Kolata 1982.
^ Kautz 2022, pp. 107-109.
^ a b Russell & Norvig 2021, p. 19.
^ a b Russell & Norvig 2021, pp. 22-23.
^ a b Kautz 2022, pp. 109-110.
^ a b c Kautz 2022, p. 110.
^ Kautz 2022, pp. 110-111.
^ a b Russell & Norvig 2021, p. 25.
^ Kautz 2022, p. 111.
^ Kautz 2020, pp. 110-111.
^ Rumelhart, David E.; Hinton, Geoffrey E.; Williams, Ronald J. (1986 ). “Learning representations by back-propagating mistakes”. Nature. 323 (6088 ): 533-536. Bibcode:1986 Natur.323..533 R. doi:10.1038/ 323533a0. ISSN 1476-4687. S2CID 205001834.
^ LeCun, Y.; Boser, B.; Denker, I.; Henderson, D.; Howard, R.; Hubbard, W.; Tackel, L. (1989 ). “Backpropagation Applied to Handwritten Postal Code Recognition”. Neural Computation. 1 (4 ): 541-551. doi:10.1162/ neco.1989.1.4.541. S2CID 41312633.
^ a b Marcus & Davis 2019.
^ a b Rossi, Francesca. “Thinking Fast and Slow in AI”. AAAI. Retrieved 5 July 2022.
^ a b Selman, Bart. “AAAI Presidential Address: The State of AI”. AAAI. Retrieved 5 July 2022.
^ a b c Kautz 2020.
^ Kautz 2022, p. 106.
^ Newell & Simon 1972.
^ & McCorduck 2004, pp. 139-179, 245-250, 322-323 (EPAM).
^ Crevier 1993, pp. 145-149.
^ McCorduck 2004, pp. 450-451.
^ Crevier 1993, pp. 258-263.
^ a b Kautz 2022, p. 108.
^ Russell & Norvig 2021, p. 9 (logicist AI), p. 19 (McCarthy’s work).
^ Maker 2006.
^ McCorduck 2004, pp. 100-101.
^ Russell & Norvig 2021, p. 2.
^ McCorduck 2004, pp. 251-259.
^ Crevier 1993, pp. 193-196.
^ Howe 1994.
^ McCorduck 2004, pp. 259-305.
^ Crevier 1993, pp. 83-102, 163-176.
^ McCorduck 2004, pp. 421-424, 486-489.
^ Crevier 1993, p. 168.
^ McCorduck 2004, p. 489.
^ Crevier 1993, pp. 239-243.
^ Russell & Norvig 2021, p. 316, 340.
^ Kautz 2022, p. 109.
^ Russell & Norvig 2021, p. 22.
^ McCorduck 2004, pp. 266-276, 298-300, 314, 421.
^ Shustek, Len (June 2010). “An interview with Ed Feigenbaum”. Communications of the ACM. 53 (6 ): 41-45. doi:10.1145/ 1743546.1743564. ISSN 0001-0782. S2CID 10239007. Retrieved 2022-07-14.
^ Lenat, Douglas B; Feigenbaum, Edward A (1988 ). “On the limits of understanding”. Proceedings of the International Workshop on Artificial Intelligence for Industrial Applications: 291-300. doi:10.1109/ AIIA.1988.13308. S2CID 11778085.
^ Russell & Norvig 2021, pp. 22-24.
^ McCorduck 2004, pp. 327-335, 434-435.
^ Crevier 1993, pp. 145-62, 197-203.
^ a b Russell & Norvig 2021, p. 23.
^ a b Clancey 1987.
^ a b Shustek, Len (2010 ). “An interview with Ed Feigenbaum”. Communications of the ACM. 53 (6 ): 41-45. doi:10.1145/ 1743546.1743564. ISSN 0001-0782. S2CID 10239007. Retrieved 2022-08-05.
^ “The fascination with AI: what is expert system?”. IONOS Digitalguide. Retrieved 2021-12-02.
^ Hayes-Roth, Murray & Adelman 2015.
^ Hayes-Roth, Barbara (1985 ). “A blackboard architecture for control”. Expert system. 26 (3 ): 251-321. doi:10.1016/ 0004-3702( 85 )90063-3.
^ Hayes-Roth, Barbara (1980 ). Human Planning Processes. RAND.
^ Pearl 1988.
^ Spiegelhalter et al. 1993.
^ Russell & Norvig 2021, pp. 335-337.
^ Russell & Norvig 2021, p. 459.
^ Quinlan, J. Ross. “Chapter 15: Learning Efficient Classification Procedures and their Application to Chess End Games”. In Michalski, Carbonell & Mitchell (1983 ).
^ Quinlan, J. Ross (1992-10-15). C4.5: Programs for Machine Learning (1st ed.). San Mateo, Calif: Morgan Kaufmann. ISBN 978-1-55860-238-0.
^ Mitchell, Tom M.; Utgoff, Paul E.; Banerji, Ranan. “Chapter 6: Learning by Experimentation: Acquiring and Refining Problem-Solving Heuristics”. In Michalski, Carbonell & Mitchell (1983 ).
^ Valiant, L. G. (1984-11-05). “A theory of the learnable”. Communications of the ACM. 27 (11 ): 1134-1142. doi:10.1145/ 1968.1972. ISSN 0001-0782. S2CID 12837541.
^ Koedinger, K. R.; Anderson, J. R.; Hadley, W. H.; Mark, M. A.; others (1997 ). “Intelligent tutoring goes to school in the big city”. International Journal of Artificial Intelligence in Education (IJAIED). 8: 30-43. Retrieved 2012-08-18.
^ Shapiro, Ehud Y (1981 ). “The Model Inference System”. Proceedings of the 7th international joint conference on Expert system. IJCAI. Vol. 2. p. 1064.
^ Manna, Zohar; Waldinger, Richard (1980-01-01). “A Deductive Approach to Program Synthesis”. ACM Trans. Program. Lang. Syst. 2 (1 ): 90-121. doi:10.1145/ 357084.357090. S2CID 14770735.
^ Schank, Roger C. (1983-01-28). Dynamic Memory: A Theory of Reminding and Learning in Computers and People. Cambridge Cambridgeshire: New York: Cambridge University Press. ISBN 978-0-521-27029-8.
^ Hammond, Kristian J. (1989-04-11). Case-Based Planning: Viewing Planning as a Memory Task. Boston: Academic Press. ISBN 978-0-12-322060-8.
^ Koza, John R. (1992-12-11). Genetic Programming: On the Programming of Computers by Means of Natural Selection (1st ed.). Cambridge, Mass: A Bradford Book. ISBN 978-0-262-11170-6.
^ Mostow, David Jack. “Chapter 12: Machine Transformation of Advice into a Heuristic Search Procedure”. In Michalski, Carbonell & Mitchell (1983 ).
^ Bareiss, Ray; Porter, Bruce; Wier, Craig. “Chapter 4: Protos: An Exemplar-Based Learning Apprentice”. In Michalski, Carbonell & Mitchell (1986 ), pp. 112-139.
^ Carbonell, Jaime. “Chapter 5: Learning by Analogy: Formulating and Generalizing Plans from Past Experience”. In Michalski, Carbonell & Mitchell (1983 ), pp. 137-162.
^ Carbonell, Jaime. “Chapter 14: Derivational Analogy: A Theory of Reconstructive Problem Solving and Expertise Acquisition”. In Michalski, Carbonell & Mitchell (1986 ), pp. 371-392.
^ Mitchell, Tom; Mabadevan, Sridbar; Steinberg, Louis. “Chapter 10: LEAP: A Knowing Apprentice for VLSI Design”. In Kodratoff & Michalski (1990 ), pp. 271-289.
^ Lenat, Douglas. “Chapter 9: The Role of Heuristics in Learning by Discovery: Three Case Studies”. In Michalski, Carbonell & Mitchell (1983 ), pp. 243-306.
^ Korf, Richard E. (1985 ). Learning to Solve Problems by Searching for Macro-Operators. Research Notes in Expert System. Pitman Publishing. ISBN 0-273-08690-1.
^ Valiant 2008.
^ a b Garcez et al. 2015.
^ Marcus 2020, p. 44.
^ Marcus 2020, p. 17.
^ a b Rossi 2022.
^ a b Selman 2022.
^ Garcez & Lamb 2020, p. 2.
^ Garcez et al. 2002.
^ Rocktäschel, Tim; Riedel, Sebastian (2016 ). “Learning Knowledge Base Inference with Neural Theorem Provers”. Proceedings of the 5th Workshop on Automated Knowledge Base Construction. San Diego, CA: Association for Computational Linguistics. pp. 45-50. doi:10.18653/ v1/W16 -1309. Retrieved 2022-08-06.
^ Serafini, Luciano; Garcez, Artur d’Avila (2016 ), Logic Tensor Networks: Deep Learning and Logical Reasoning from Data and Knowledge, arXiv:1606.04422.
^ a b Garcez, Artur d’Avila; Lamb, Luis C.; Gabbay, Dov M. (2009 ). Neural-Symbolic Cognitive Reasoning (1st ed.). Berlin-Heidelberg: Springer. Bibcode:2009 nscr.book … D. doi:10.1007/ 978-3-540-73246-4. ISBN 978-3-540-73245-7. S2CID 14002173.
^ Kiczales, Gregor; Rivieres, Jim des; Bobrow, Daniel G. (1991-07-30). The Art of the Metaobject Protocol (1st ed.). Cambridge, Mass: The MIT Press. ISBN 978-0-262-61074-2.
^ Motik, Boris; Shearer, Rob; Horrocks, Ian (2009-10-28). “Hypertableau Reasoning for Description Logics”. Journal of Expert System Research. 36: 165-228. arXiv:1401.3485. doi:10.1613/ jair.2811. ISSN 1076-9757. S2CID 190609.
^ Kuipers, Benjamin (1994 ). Qualitative Reasoning: Modeling and Simulation with Incomplete Knowledge. MIT Press. ISBN 978-0-262-51540-5.
^ Russell & Norvig 2021.
^ Leo de Penning, Artur S. d’Avila Garcez, Luís C. Lamb, John-Jules Ch. Meyer: “A Neural-Symbolic Cognitive Agent for Online Learning and Reasoning.” IJCAI 2011: 1653-1658.
^ McCarthy & Hayes 1969.
^ McCarthy 1959.
^ Nilsson 1998, p. 7.
^ Olazaran 1993, pp. 411-416.
^ Olazaran 1993, pp. 415-416.
^ Marcus 2020, p. 20.
^ Garcez & Lamb 2020, p. 8.
^ a b Russell & Norvig 2021, p. 982.
^ Brooks 1991, p. 143.
^ Brooks 1991, p. 151.
^ Brooks 1991, p. 150.
^ Brooks 1991, p. 142.
References

Brooks, Rodney A. (1991 ). “Intelligence without representation”. Expert system. 47 (1 ): 139-159. doi:10.1016/ 0004-3702( 91 )90053-M. ISSN 0004-3702. S2CID 207507849. Retrieved 2022-09-13.
Clancey, William (1987 ). Knowledge-Based Tutoring: The GUIDON Program (MIT Press Series in Artificial Intelligence) (Hardcover ed.).
Crevier, Daniel (1993 ). AI: The Tumultuous Look For Artificial Intelligence. New York City, NY: BasicBooks. ISBN 0-465-02997-3.
Dreyfus, Hubert L (1981 ). “From micro-worlds to understanding representation: AI at an impasse” (PDF). Mind Design. MIT Press, Cambridge, MA: 161-204.
Garcez, Artur S. d’Avila; Broda, Krysia; Gabbay, Dov M.; Gabbay, Augustus de Morgan Professor of Logic Dov M. (2002 ). Neural-Symbolic Learning Systems: Foundations and Applications. Springer Science & Business Media. ISBN 978-1-85233-512-0.
Garcez, Artur; Besold, Tarek; De Raedt, Luc; Földiák, Peter; Hitzler, Pascal; Icard, Thomas; Kühnberger, Kai-Uwe; Lamb, Luís; Miikkulainen, Risto; Silver, Daniel (2015 ). Neural-Symbolic Learning and Reasoning: Contributions and Challenges. AAI Spring Symposium – Knowledge Representation and Reasoning: Integrating Symbolic and Neural Approaches. Stanford, CA: AAAI Press. doi:10.13140/ 2.1.1779.4243.
Garcez, Artur d’Avila; Gori, Marco; Lamb, Luis C.; Serafini, Luciano; Spranger, Michael; Tran, Son N. (2019 ), Neural-Symbolic Computing: An Effective Methodology for Principled Integration of Artificial Intelligence and Reasoning, arXiv:1905.06088.
Garcez, Artur d’Avila; Lamb, Luis C. (2020 ), Neurosymbolic AI: The 3rd Wave, arXiv:2012.05876.
Haugeland, John (1985 ), Artificial Intelligence: The Very Idea, Cambridge, Mass: MIT Press, ISBN 0-262-08153-9.
Hayes-Roth, Frederick; Murray, William; Adelman, Leonard (2015 ). “Expert systems”. AccessScience. doi:10.1036/ 1097-8542.248550.
Honavar, Vasant; Uhr, Leonard (1994 ). Symbolic Artificial Intelligence, Connectionist Networks & Beyond (Technical report). Iowa State University Digital Repository, Computer Science Technical Reports. 76. p. 6.
Honavar, Vasant (1995 ). Symbolic Artificial Intelligence and Numeric Artificial Neural Networks: Towards a Resolution of the Dichotomy. The Springer International Series In Engineering and Computer Science. Springer US. pp. 351-388. doi:10.1007/ 978-0-585-29599-2_11.
Howe, J. (November 1994). “Artificial Intelligence at Edinburgh University: a Viewpoint”. Archived from the original on 15 May 2007. Retrieved 30 August 2007.
Kautz, Henry (2020-02-11). The Third AI Summer, Henry Kautz, AAAI 2020 Robert S. Engelmore Memorial Award Lecture. Retrieved 2022-07-06.
Kautz, Henry (2022 ). “The Third AI Summer: AAAI Robert S. Engelmore Memorial Lecture”. AI Magazine. 43 (1 ): 93-104. doi:10.1609/ aimag.v43i1.19122. ISSN 2371-9621. S2CID 248213051. Retrieved 2022-07-12.
Kodratoff, Yves; Michalski, Ryszard, eds. (1990 ). Machine Learning: an Artificial Intelligence Approach. Vol. III. San Mateo, Calif.: Morgan Kaufman. ISBN 0-934613-09-5. OCLC 893488404.
Kolata, G. (1982 ). “How can computers get good sense?”. Science. 217 (4566 ): 1237-1238. Bibcode:1982 Sci … 217.1237 K. doi:10.1126/ science.217.4566.1237. PMID 17837639.
Maker, Meg Houston (2006 ). “AI@50: AI Past, Present, Future”. Dartmouth College. Archived from the initial on 3 January 2007. Retrieved 16 October 2008.
Marcus, Gary; Davis, Ernest (2019 ). Rebooting AI: Building Artificial Intelligence We Can Trust. New York: Pantheon Books. ISBN 9781524748258. OCLC 1083223029.
Marcus, Gary (2020 ), The Next Decade in AI: Four Steps Towards Robust Expert system, arXiv:2002.06177.
McCarthy, John (1959 ). PROGRAMS WITH COMMON SENSE. Symposium on Mechanization of Thought Processes. NATIONAL PHYSICAL LABORATORY, TEDDINGTON, UK. p. 8.
McCarthy, John; Hayes, Patrick (1969 ). “Some Philosophical Problems From the Standpoint of Expert System”. Machine Intelligence 4. B. Meltzer, Donald Michie (eds.): 463-502.
McCorduck, Pamela (2004 ), Machines Who Think (2nd ed.), Natick, Massachusetts: A. K. Peters, ISBN 1-5688-1205-1.
Michalski, Ryszard; Carbonell, Jaime; Mitchell, Tom, eds. (1983 ). Artificial intelligence: an Artificial Intelligence Approach. Vol. I. Palo Alto, Calif.: Tioga Publishing Company. ISBN 0-935382-05-4. OCLC 9262069.
Michalski, Ryszard; Carbonell, Jaime; Mitchell, Tom, eds. (1986 ). Artificial intelligence: an Expert System Approach. Vol. II. Los Altos, Calif.: Morgan Kaufman. ISBN 0-934613-00-1.
Newell, Allen; Simon, Herbert A. (1972 ). Human Problem Solving (1st ed.). Englewood Cliffs, New Jersey: Prentice Hall. ISBN 0-13-445403-0.
Newell, Allen; Simon, H. A. (1976 ). “Computer Science as Empirical Inquiry: Symbols and Search”. Communications of the ACM. 19 (3 ): 113-126. doi:10.1145/ 360018.360022.
Nilsson, Nils (1998 ). Artificial Intelligence: A New Synthesis. Morgan Kaufmann. ISBN 978-1-55860-467-4. Archived from the original on 26 July 2020. Retrieved 18 November 2019.
Olazaran, Mikel (1993-01-01), “A Sociological History of the Neural Network Controversy”, in Yovits, Marshall C. (ed.), Advances in Computers Volume 37, vol. 37, Elsevier, pp. 335-425, doi:10.1016/ S0065-2458( 08 )60408-8, ISBN 9780120121373, obtained 2023-10-31.
Pearl, J. (1988 ). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. San Mateo, California: Morgan Kaufmann. ISBN 978-1-55860-479-7. OCLC 249625842.
Russell, Stuart J.; Norvig, Peter (2021 ). Expert system: A Modern Approach (fourth ed.). Hoboken: Pearson. ISBN 978-0-13-461099-3. LCCN 20190474.
Rossi, Francesca (2022-07-06). “AAAI2022: Thinking Fast and Slow in AI (AAAI 2022 Invited Talk)”. Retrieved 2022-07-06.
Selman, Bart (2022-07-06). “AAAI2022: Presidential Address: The State of AI”. Retrieved 2022-07-06.
Serafini, Luciano; Garcez, Artur d’Avila (2016-07-07), Logic Tensor Networks: Deep Learning and Logical Reasoning from Data and Knowledge, arXiv:1606.04422.
Spiegelhalter, David J.; Dawid, A. Philip; Lauritzen, Steffen; Cowell, Robert G. (1993 ). “Bayesian analysis in specialist systems”. Statistical Science. 8 (3 ).
Turing, A. M. (1950 ). “I.-Computing Machinery and Intelligence”. Mind. LIX (236 ): 433-460. doi:10.1093/ mind/LIX.236.433. ISSN 0026-4423. Retrieved 2022-09-14.
Valiant, Leslie G (2008 ). “Knowledge Infusion: In Pursuit of Robustness in Expert System”. In Hariharan, R.; Mukund, M.; Vinay, V. (eds.). Foundations of Software Technology and Theoretical Computer Science (Bangalore). pp. 415-422.
Xifan Yao; Jiajun Zhou; Jiangming Zhang; Claudio R. Boer (2017 ). From Intelligent Manufacturing to Smart Manufacturing for Industry 4.0 Driven by Next Generation Artificial Intelligence and Further On.