The Future is Neuro-Symbolic: How AI Reasoning is Evolving by Anthony Alcaraz

If the capacity for symbolic reasoning is in fact idiosyncratic and context-dependent in the way suggested here, what are the implications for scientific psychology? Therefore, the key to understanding the human capacity for symbolic reasoning in general will be to characterize typical sensorimotor strategies, and to understand the particular conditions in which those strategies are successful or unsuccessful. Hinton and many others have tried hard to banish symbols altogether. The deep learning hope—seemingly grounded not so much in science, but in a sort of historical grudge—is that intelligent behavior will emerge purely from the confluence of massive data and deep learning. A key component of the system architecture for all expert systems is the knowledge base, which stores facts and rules for problem-solving.[51]

The simplest approach for an expert system knowledge base is simply a collection or network of production rules.

Like interlocking puzzle pieces that together form a larger image, sensorimotor mechanisms and physical notations “interlock” to produce sophisticated mathematical behaviors. Insofar as mathematical rule-following emerges from active engagement with physical notations, the mathematical rule-follower is a distributed system that spans the boundaries between brain, body, and environment. For this interlocking to promote mathematically appropriate behavior, however, the relevant perceptual and sensorimotor mechanisms must be just as well-trained as the physical notations must be well-designed. Thus, on one hand, the development of symbolic reasoning abilities in an individual subject will depend on the development of a sophisticated sensorimotor skillset in the way outlined above. While the particular techniques in symbolic AI varied greatly, the field was largely based on mathematical logic, which was seen as the proper (“neat”) representation formalism for most of the underlying concepts of symbol manipulation.

When you provide it with a new image, it will return the probability that it contains a cat. There have been several efforts to create complicated symbolic AI systems that encompass the multitudes of rules of certain domains. Called expert systems, these symbolic AI models use hardcoded knowledge and rules to tackle complicated tasks such as medical diagnosis. But they require a huge amount of effort by domain experts and software engineers and only work in very narrow use cases. As soon as you generalize the problem, there will be an explosion of new rules to add (remember the cat detection problem?), which will require more human labor.

The idea was based on the, now commonly exemplified, fact that logical connectives of conjunction and disjunction can be easily encoded by binary threshold units with weights — i.e., the perceptron, an elegant learning algorithm for which was introduced shortly. However, given the aforementioned recent evolution of the neural/deep learning concept, the NSI field is now gaining more momentum than ever. Once they are built, symbolic methods tend to be faster and more efficient than neural techniques.

Thinking correctly and effectively requires training in Logic, just as writing well requires training in English and composition. Without explicit training, we are likely to be unsure of our conclusions; we are prone to make mistakes; and we are apt to be fooled by others. P.J.B. performed the research, contributed new analytical tools and analyzed data. “Pushing symbols,” Proceedings of the 31st Annual Conference of the Cognitive Science Society. To think that we can simply abandon symbol-manipulation is to suspend disbelief.

Data Dependency:

“Having language models reason with code unlocks many opportunities for tool use, output validation, more structured understanding into model’s capabilities and way of thinking, and more,” says Leonid Karlinsky, principal scientist at the MIT-IBM what is symbolic reasoning Watson AI Lab. The approach also offers greater efficiency than some other methods. If a user has many similar questions, they can generate one core program and then replace certain variables without needing to run the model repeatedly.

That’s not effective, either—the whole point of symbolism is for it to communicate with readers at a level beyond the literal, acting almost like a form of subliminal messaging. In some cases, symbolism is broad and used to communicate a work’s theme, like Aslan the lion in The Lion, the Witch and the Wardrobe as a symbol of Christ. In other cases, symbolism is used to communicate details about a character, setting, or plot point, such as a black cat being used to symbolize a character’s bad luck.

As opposed to pure neural network–based models, the hybrid AI can learn new tasks with less data and is explainable. And unlike symbolic-only models, NSCL doesn’t struggle to analyze the content of images. Parsing, tokenizing, spelling correction, part-of-speech tagging, noun and verb phrase chunking are all aspects of natural language processing long handled by symbolic AI, but since improved by deep learning approaches.

Expert systems are monotonic; that is, the more rules you add, the more knowledge is encoded in the system, but additional rules can’t undo old knowledge. Monotonic basically means one direction; i.e. when one thing goes up, another thing goes up. Because machine learning algorithms can be retrained on new data, and will revise their parameters based on that new data, they are better at encoding tentative knowledge that can be retracted later if necessary. Because machine learning algorithms can be retrained on new data, and will revise their parameters based on that new data, they are better at encoding tentative knowledge that can be retracted later if necessary; i.e. if they need to learn something new, like when data is non-stationary.

How to boost language models with graph neural networks

Knowledge-based systems have an explicit knowledge base, typically of rules, to enhance reusability across domains by separating procedural code and domain knowledge. A separate inference engine processes rules and adds, deletes, or modifies a knowledge store. Multiple different approaches to represent knowledge and then reason with those representations have been investigated. Below is a quick overview of approaches to knowledge representation and automated reasoning. During the first AI summer, many people thought that machine intelligence could be achieved in just a few years.

Japan championed Prolog for its Fifth Generation Project, intending to build special hardware for high performance. Similarly, LISP machines were built to run LISP, but as the second AI boom turned to bust these companies could not compete with new workstations that could now run LISP or Prolog natively at comparable speeds. This section provides an overview of techniques and contributions in an overall context leading to many other, more detailed articles in Wikipedia. You can foun additiona information about ai customer service and artificial intelligence and NLP. Sections on Machine Learning and Uncertain Reasoning are covered earlier in the history section. Time periods and titles are drawn from Henry Kautz’s 2020 AAAI Robert S. Engelmore Memorial Lecture[17] and the longer Wikipedia article on the History of AI, with dates and titles differing slightly for increased clarity. The words sign and symbol derive from Latin and Greek words, respectively, that mean mark or token, as in “take this rose as a token of my esteem.” Both words mean “to stand for something else” or “to represent something else”.

The combination of neural and symbolic approaches has reignited a long-simmering debate in the AI community about the relative merits of symbolic approaches (e.g., if-then statements, decision trees, mathematics) and neural approaches (e.g., deep learning and, more recently, generative AI). We see Neuro-symbolic AI as a pathway to achieve artificial general intelligence. By augmenting and combining the strengths of statistical AI, like machine learning, with the capabilities of human-like symbolic knowledge and reasoning, we’re aiming to create a revolution in AI, rather than an evolution. One is based on possible worlds; the other is based on symbolic manipulation of expressions. Yet, for “well-behaved” logics, it turns out that logical entailment and provability are identical – a set of premises logically entails a conclusion if and only if the conclusion is provable from the premises. Even if the number of worlds is infinite, it is possible in such logics to produce a finite proof of the conclusion, i.e. we can determine logical entailment without going through all possible worlds.

Most of the existing literature on symbolic reasoning has been developed using an implicitly or explicitly translational perspective. Although we do not believe that the current evidence is enough to completely dislodge this perspective, it does show that sensorimotor processing influences the capacity for symbolic reasoning in a number of interesting and surprising ways. The translational view easily accounts for cases in which individual symbols are more readily perceived based on external format. Perceptual Manipulations Theory also predicts this sort of impact, but further predicts that perceived structures will affect the application of rules—since rules are presumed to be implemented via systems involved in perceiving that structure. In this section, we will review several empirical sources of evidence for the impact of visual structure on the implementation of formal rules. Although translational accounts may eventually be elaborated to accommodate this evidence, it is far more easily and naturally accommodated by accounts which, like PMT, attribute a constitutive role to perceptual processing.

Deduction is a form of symbolic reasoning that produces conclusions that are logically entailed by premises (distinguishing it from other forms of reasoning, such as induction, abduction, and analogical reasoning). A proof is a sequence of simple, more-or-less obvious deductive steps that justifies a conclusion that may not be immediately obvious from given premises. In Logic, we usually encode logical information as sentences in formal languages; and we use rules of inference appropriate to these languages.

The course presumes that the student understands sets and set operations, such as union, intersection, and complement. The course also presumes that the student is comfortable with symbolic mathematics, at the level of high-school algebra. However, it has been used by motivated secondary school students and post-graduate professionals interested in honing their logical reasoning skills. As ‘common sense’ AI matures, it will be possible to use it for better customer support, business intelligence, medical informatics, advanced discovery, and much more. The universe is written in the language of mathematics and its characters are triangles, circles, and other geometric objects. One solution is to take pictures of your cat from different angles and create new rules for your application to compare each input against all those images.

Say you have a picture of your cat and want to create a program that can detect images that contain your cat. You create a rule-based program that takes new images as inputs, compares the pixels to the original cat image, and responds by saying whether your cat is in those images. Using OOP, you can create extensive and complex symbolic AI programs that perform various tasks. Many of the concepts and tools you find in computer science are the results of these efforts. Symbolic AI programs are based on creating explicit structures and behavior rules.

This page includes some recent, notable research that attempts to combine deep learning with symbolic learning to answer those questions. Symbolic Artificial Intelligence continues to be a vital part of AI research and applications. Its ability to process and apply complex sets of rules and logic makes it indispensable in various domains, complementing other AI methodologies like Machine Learning and Deep Learning.

We compare Schema Networks with Asynchronous Advantage Actor-Critic and Progressive Networks on a suite of Breakout variations, reporting results on training efficiency and zero-shot generalization, consistently demonstrating faster, more robust learning and better transfer. We argue that generalizing from limited data and learning causal relationships are essential abilities on the path toward generally intelligent systems. Combining symbolic reasoning with deep neural networks and deep reinforcement learning may help us address the fundamental challenges of reasoning, hierarchical representations, transfer learning, robustness in the face of adversarial examples, and interpretability (or explanatory power). We have described an approach to symbolic reasoning which closely ties it to the perceptual and sensorimotor mechanisms that engage physical notations.

There are now several efforts to combine neural networks and symbolic AI. One such project is the Neuro-Symbolic Concept Learner (NSCL), a hybrid AI system developed by the MIT-IBM Watson AI Lab. NSCL uses both rule-based programs and neural networks to solve visual question-answering problems.

There are 216 (65,536) possible combinations of these true-false possibilities, and so there are 216 possible worlds. It is used primarily by mathematicians in proving complicated theorems in geometry or number theory. It is all about writing formal proofs to be published in scholarly papers that have little to do with everyday life. Logic is important in all of these disciplines, and it is essential in computer science.

This approach provides interpretability, generalizability, and robustness— all critical requirements in enterprise NLP settings . Deep neural networks are also very suitable for reinforcement learning, AI models that develop their behavior through numerous trial and error. This is the kind of AI that masters complicated games such as Go, StarCraft, and Dota. Symbolic artificial intelligence showed early progress at the dawn of AI and computing. You can easily visualize the logic of rule-based programs, communicate them, and troubleshoot them.

This book provides a broad overview of the key results and frameworks for various NSAI tasks as well as discussing important application areas. This book also covers neuro symbolic reasoning frameworks such as LNN, LTN, and NeurASP and learning frameworks. This would include differential inductive logic programming, constraint learning and deep symbolic policy learning.

Nevertheless, there is probably no uniquely correct answer to the question of how people do mathematics. Indeed, it is important to consider the relative merits of all competing accounts and to incorporate the best elements of each. Although we believe that most of our mathematical abilities are rooted in our past experience and engagement with notations, we do not depend on these notations at all times.

Google announced a new architecture for scaling neural network architecture across a computer cluster to train deep learning algorithms, leading to more innovation in neural networks. AI neural networks are modeled after the statistical properties of interconnected neurons in the human brain and brains of other animals. These artificial neural networks (ANNs) create a framework for modeling patterns in data represented by slight changes in the connections between individual neurons, which in turn enables the neural network to keep learning and picking out patterns in data. This can help tease apart features at different levels of abstraction. In the case of images, this could include identifying features such as edges, shapes and objects.

While neuro symbolic ideas date back to the early 2000’s, there have been significant advances in the last five years. Common symbolic AI algorithms include expert systems, logic programming, semantic networks, Bayesian networks and fuzzy logic. These algorithms are used for knowledge representation, reasoning, planning and decision-making. They work well for applications with well-defined workflows, but struggle when apps are trying to make sense of edge cases.

In what follows, we articulate a constitutive account of symbolic reasoning, Perceptual Manipulations Theory, that seeks to elaborate on the cyborg view in exactly this way. On our view, the way in which physical notations are perceived is at least as important as the way in which they are actively manipulated. This book is designed for researchers and advanced-level students trying to understand the current landscape of NSAI research as well as those looking to apply NSAI research in areas such as natural language processing and visual question answering. Practitioners who specialize in employing machine learning and AI systems for operational use will find this book useful as well. And unlike symbolic AI, neural networks have no notion of symbols and hierarchical representation of knowledge. This limitation makes it very hard to apply neural networks to tasks that require logic and reasoning, such as science and high-school math.

On our view, therefore, much of the capacity for symbolic reasoning is implemented as the perception, manipulation and modal and cross-modal representation of externally perceived notations. Analogous to the syntactic approach above, computationalism holds that the capacity for symbolic reasoning is carried out by mental processes of syntactic rule-based symbol-manipulation. In its canonical form, these processes take place in a general-purpose “central reasoning system” that is functionally encapsulated from dedicated and modality-specific sensorimotor “modules” (Fodor, 1983; Sloman, 1996; Pylyshyn, 1999; Anderson, 2007).

The second says that m and r implies p or q, i.e. if it is Monday and raining, then Mary loves Pat or Mary loves Quincy. As an illustration of errors that arise in reasoning with sentences in natural language, consider the following examples. https://chat.openai.com/ In the first, we use the transitivity of the better relation to derive a conclusion about the relative quality of champagne and soda from the relative quality of champagne and beer and the relative quality or beer and soda.

Functional Logic takes us one step further by providing a means for describing worlds with infinitely many objects. The resulting logic is much more powerful than Propositional Logic and Relational Logic. Unfortunately, as we shall see, some of the nice computational properties of the first two logics are lost as a result.

And for the final step, the model outputs the result as a line of natural language with an automatic data visualization, if needed. “We want AI to perform complex reasoning in a way that is transparent and trustworthy. While the aforementioned correspondence between the propositional logic formulae and neural networks has been very direct, transferring the same principle to the relational setting was a major challenge NSI researchers have been traditionally struggling with. The issue is that in the propositional setting, only the (binary) values of the existing input propositions are changing, with the structure of the logical program being fixed. Driven heavily by the empirical success, DL then largely moved away from the original biological brain-inspired models of perceptual intelligence to “whatever works in practice” kind of engineering approach.

As computational capacities grow, the way we digitize and process our analog reality can also expand, until we are juggling billion-parameter tensors instead of seven-character strings. Symbolism is the use of words or images to symbolize specific concepts, people, objects, or events. The key here is that the symbols used aren’t literal representations, but figurative or implied ones. For example, starting a personal essay about transformation with imagery of a butterfly. It wasn’t until the 1980’s, when the chain rule for differentiation of nested functions was introduced as the backpropagation method to calculate gradients in such neural networks which, in turn, could be trained by gradient descent methods.

On one hand, students can think about such problems syntactically, as a specific instance of the more general logical form “All Xs are Ys; All Ys are Zs; Therefore, all Xs are Zs.” On the other hand, they might think about them semantically—as relations between subsets, for example. In an analogous fashion, two prominent scientific attempts to explain how students are able to solve symbolic reasoning problems can be distinguished according to their emphasis on syntactic or semantic properties. A certain set of structural rules are innate to humans, independent of sensory experience.

Including symbolism in your writing doesn’t mean you have to “swap out” literal descriptions; it often enhances these literal descriptions. You can recognize symbolism when an image in a piece of text seems to indicate something other than its literal meaning. It might be repeated or seem somewhat jarring, as if the author is intentionally pointing it out (and they might be—though authors don’t always Chat GPT do this). For example, a character might be described as having piercing green eyes that fixate on others. Symbolism can be obvious to the point of feeling too obvious, like naming an evil character Nick DeVille and describing his hairstyle as being reminiscent of horns. When this is the case, you might only recognize the symbolism on a second read-through, once you know how the story ends.

• This concept is fundamental in AI Research Labs and universities, contributing to significant Development Milestones in AI.
• We say that a set of premises logically entails a conclusion if and only if every world that satisfies the premises also satisfies the conclusion.
• This approach provides interpretability, generalizability, and robustness— all critical requirements in enterprise NLP settings .
• On one hand, students can think about such problems syntactically, as a specific instance of the more general logical form “All Xs are Ys; All Ys are Zs; Therefore, all Xs are Zs.” On the other hand, they might think about them semantically—as relations between subsets, for example.

We can think of individual reasoning steps as the atoms out of which proof molecules are built. By writing logical sentences, each informant can express exactly what he or she knows – no more, no less. The following sentences are examples of different types of logical sentences. The first sentence is straightforward; it tells us directly that Dana likes Cody. The second and third sentences tell us what is not true without saying what is true.

Deep learning and neuro-symbolic AI 2011–now

The work in AI started by projects like the General Problem Solver and other rule-based reasoning systems like Logic Theorist became the foundation for almost 40 years of research. Symbolic AI (or Classical AI) is the branch of artificial intelligence research that concerns itself with attempting to explicitly represent human knowledge in a declarative form (i.e. facts and rules). If such an approach is to be successful in producing human-like intelligence then it is necessary to translate often implicit or procedural knowledge possessed by humans into an explicit form using symbols and rules for their manipulation. Artificial systems mimicking human expertise such as Expert Systems are emerging in a variety of fields that constitute narrow but deep knowledge domains. According to Wikipedia, machine learning is an application of artificial intelligence where “algorithms and statistical models are used by computer systems to perform a specific task without using explicit instructions, relying on patterns and inference instead. (…) Machine learning algorithms build a mathematical model based on sample data, known as ‘training data’, in order to make predictions or decisions without being explicitly programmed to perform the task”.

Early deep learning systems focused on simple classification tasks like recognizing cats in videos or categorizing animals in images. However, innovations in GenAI techniques such as transformers, autoencoders and generative adversarial networks have opened up a variety of use cases for using generative AI to transform unstructured data into more useful structures for symbolic processing. Now, researchers are looking at how to integrate these two approaches at a more granular level for discovering proteins, discerning business processes and reasoning. Over the next few decades, research dollars flowed into symbolic methods used in expert systems, knowledge representation, game playing and logical reasoning.

This way of using rules in AI has been around for a long time and is really important for understanding how computers can be smart. It’s represented various causes and sentiments over the country’s history and in the wake of the Jan. 6 insurrection at the U.S. Symbolism isn’t just something you find in literature; it’s found in architecture, city planning, historical events, and just about every other area of life. For example, NASA’s Apollo missions, the series of missions that landed the first humans on the moon, were named for the Greek god Apollo.

The next wave of innovation will involve combining both techniques more granularly. Both symbolic and neural network approaches date back to the earliest days of AI in the 1950s. On the symbolic side, the Logic Theorist program in 1956 helped solve simple theorems. The Perceptron algorithm in 1958 could recognize simple patterns on the neural network side. However, neural networks fell out of favor in 1969 after AI pioneers Marvin Minsky and Seymour Papert published a paper criticizing their ability to learn and solve complex problems. Popular categories of ANNs include convolutional neural networks (CNNs), recurrent neural networks (RNNs) and transformers.

Unlike many traditional accounts, PMT does not presuppose that mathematical and logical rules must be internally represented in order to be followed. Logic is the study of information encoded in the form of logical sentences. Each logical sentence divides the set of all possible world into two subsets – the set of worlds in which the sentence is true and the set of worlds in which the set of sentences is false. A set of premises logically entails a conclusion if and only if the conclusion is true in every world in which all of the premises are true.

For that, however, researchers had to replace the originally used binary threshold units with differentiable activation functions, such as the sigmoids, which started digging a gap between the neural networks and their crisp logical interpretations. This only escalated with the arrival of the deep learning (DL) era, with which the field got completely dominated by the sub-symbolic, continuous, distributed representations, seemingly ending the story of symbolic AI. Amongst the main advantages of this logic-based approach towards ML have been the transparency to humans, deductive reasoning, inclusion of expert knowledge, and structured generalization from small data. And while the current success and adoption of deep learning largely overshadowed the preceding techniques, these still have some interesting capabilities to offer.

This example is interesting in that it showcases our formal language for encoding logical information. As with algebra, we use symbols to represent relevant aspects of the world in question, and we use operators to connect these symbols in order to express information about the things those symbols represent. First of all, correctness in logical reasoning is determined by the logical operators in our sentences, not the objects and relationships mentioned in those sentences. Second, the conclusion is guaranteed to be true only if the premises are true. In this work, we approach KBQA with the basic premise that if we can correctly translate the natural language questions into an abstract form that captures the question’s conceptual meaning, we can reason over existing knowledge to answer complex questions. Table 1 illustrates the kinds of questions NSQA can handle and the form of reasoning required to answer different questions.

Therefore, symbols have also played a crucial role in the creation of artificial intelligence. We use symbols all the time to define things (cat, car, airplane, etc.) and people (teacher, police, salesperson). Symbols can represent abstract concepts (bank transaction) or things that don’t physically exist (web page, blog post, etc.).

The primary operators are Boolean connectives, such as and, or, and not. The language of Logic can be used to encode regulations and business rules, and automated reasoning techniques can be used to analyze such regulations for inconsistency and overlap. Logical spreadsheets generalize traditional spreadsheets to include logical constraints as well as traditional arithmetic formulas. For example, in scheduling applications, we might have timing constraints or restrictions on who can reserve which rooms. In the domain of travel reservations, we might have constraints on adults and infants. In academic program sheets, we might have constraints on how many courses of varying types that students must take.

Neuro symbolic reasoning and learning is a topic that combines ideas from deep neural networks with symbolic reasoning and learning to overcome several significant technical hurdles such as explainability, modularity, verification, and the enforcement of constraints. While neuro symbolic ideas date back to the early 2000’s, there have been significant advances in the last 5 years. In this chapter, we outline some of these advancements and discuss how they align with several taxonomies for neuro symbolic reasoning. Neuro symbolic AI is a topic that combines ideas from deep neural networks with symbolic reasoning and learning to overcome several significant technical hurdles such as explainability, modularity, verification, and the enforcement of constraints.

Instead, they produce task-specific vectors where the meaning of the vector components is opaque. We investigate an unconventional direction of research that aims at converting neural networks, a class of distributed, connectionist, sub-symbolic models into a symbolic level with the ultimate goal of achieving AI interpretability and safety. To that end, we propose Object-Oriented Deep Learning, a novel computational paradigm of deep learning that adopts interpretable “objects/symbols” as a basic representational atom instead of N-dimensional tensors (as in traditional “feature-oriented” deep learning). For visual processing, each “object/symbol” can explicitly package common properties of visual objects like its position, pose, scale, probability of being an object, pointers to parts, etc., providing a full spectrum of interpretable visual knowledge throughout all layers.

The AMR is aligned to the terms used in the knowledge graph using entity linking and relation linking modules and is then transformed to a logic representation.5 This logic representation is submitted to the LNN. LNN performs necessary reasoning such as type-based and geographic reasoning to eventually return the answers for the given question. For example, Figure 3 shows the steps of geographic reasoning performed by LNN using manually encoded axioms and DBpedia Knowledge Graph to return an answer.