Beyond the symbols vs signals debate

We seek to define the capability that has enabled humans to develop the civilisation we have, and that distinguishes us from other species. For this it is not enough to identify a distinguishing characteristic - we want a capability that is also explanatory of humanity's achievements. "Intelligence" does not work here because we have no agreed definition of what intelligence is or how an intelligent entity behaves. We need a concept that is behaviourally better defined. The definition will need to be computational in the sense that the expected outcomes of exercising the capability need to be both specifiable and computationally feasible. This formulation is related to the goals of AI research but is not synonymous with it, leaving out the many capabilities we share with other species.

We make a proposal for this essential human capability, which we call "educability." It synthesises abilities to learn from experience, to learn from others, and to chain together what we have learned in either mode and apply that to particular situations. It starts with the now standard notion of learning from examples, as captured by the Probably Approximately Correct model and used in machine learning. The ability of Large Language Models learning from examples to generate smoothly flowing prose lends encouragement to this approach. The basic question then is how to extend this approach to encompass broader human capabilities beyond learning from examples. This is what the educability notion aims to answer.

Professor Leslie Valiant FRS< Harvard University, USA

Professor Leslie Valiant FRS< Harvard University, USA

Leslie Valiant was educated at King's College, Cambridge; Imperial College, London; and Warwick University where he received his PhD in computer science in 1974. He is currently T Jefferson Coolidge Professor of Computer Science and Applied Mathematics in the Division of Engineering and Applied Sciences at Harvard, where he has taught since 1982. Before coming to Harvard, he had taught at Carnegie-Mellon University, Leeds University, and the University of Edinburgh.

His work has ranged over several areas of theoretical computer science, particularly complexity theory, computational learning, and parallel computation. He also has interests in computational neuroscience, evolution and artificial intelligence.

He received the Nevanlinna Prize at the International Congress of Mathematicians in 1986; the Knuth Award in 1997; the EATCS award in 2008; and the ACM AM Turing Award in 2010. He is a Fellow of the Royal Society (London) and a member of the National Academy of Sciences (USA).

Neurosymbolic AI (NeSy) is regarded as the third wave in AI. It aims at combining knowledge representation and reasoning with neural networks. Numerous approaches to NeSy are being developed and there exists an `alphabet-soup' of different systems, whose relationships are often unclear. I will discuss the state-of-the art in NeSy and argue that there are many similarities with statistical relational AI (StarAI).

Taking inspiring from StarAI, and exploiting these similarities, I will argue that Neurosymbolic AI = Logic + Probability + Neural Networks.  I will also provide a recipe for developing NeSy approaches: start from a logic, add a probabilistic interpretation, and then turn neural networks into `neural predicates'.

Probability is interpreted broadly here, and is necessary to provide a quantitative and differentiable component to the logic. At the semantic and the computation level, one can then combine logical circuits (aka proof structures) labelled with probability, and neural networks in computation graphs. 

I will illustrate the recipe with NeSy systems such as DeepProbLog, a deep probabilistic extension of Prolog, and DeepStochLog, a neural network extension of stochastic definite clause grammars (or stochastic logic programs).

Professor Luc De Raedt, KU Leuven and Örebro University, Belgium

Professor Luc De Raedt, KU Leuven and Örebro University, Belgium

Professor Dr Luc De Raedt is Director of Leuven.AI, the KU Leuven Institute for AI, full professor of Computer Science at KU Leuven, and guest professor at Örebro University (Sweden) at the Center for Applied Autonomous Sensor Systems in the Wallenberg AI, Autonomous Systems and Software Program. He is working on the integration of machine learning and machine reasoning techniques, also known under the term neurosymbolic AI.  He has chaired the main European and International Machine Learning and Artificial Intelligence conferences (IJCAI, ECAI, ICML and ECMLPKDD) and is a fellow of EurAI,  AAAI and ELLIS, and member of Royal Flemish Academy of Belgium. He received ERC Advanced Grants in 2015 and 2023.

What do we need to build artificial agents which can reason effectively and generalise to new situations? An oft-cited claim, both in cognitive science and in machine learning, is that a key ingredient for reasoning and generalisation is planning with a model of the world. In this talk, Dr Hamrick will evaluate this claim in the context of model-based reinforcement learning, presenting evidence that demonstrates the utility of planning for certain classes of problems (e.g. in-distribution learning and procedural generalisation in reinforcement learning), as well as evidence that planning is not a silver bullet for out-of-distribution generalisation. In particular, generalisation performance is limited by the generalisation abilities of the individual components required for planning (e.g., the policy, reward model, and world model), which in turn are dependent on the diversity of data those components are trained on. Moreover, generalisation is strongly dependent on choosing the appropriate level of abstraction. These concerns may be partially addressed by leveraging new state-of-the-art foundation models, which are trained on both an unprecedented breadth of data and at a higher level of abstraction than before.

Dr Jessica Hamrick, Google DeepMind, UK

Dr Jessica Hamrick, Google DeepMind, UK

Dr Hamrick is a Staff Research Scientist at Google DeepMind where she co-leads the PRISM (Planning, Reasoning, Inference & Structured Models) team. Her current work focuses on improving the reasoning capabilities of large language models, and she has previously worked on topics spanning model-based reinforcement learning, planning, graph neural networks, and computational modelling of mental simulation. Dr Hamrick received her PhD (2017) in Psychology from the University of California, Berkeley and her BS and MEng (2012) in Computer Science from the Massachusetts Institute of Technology. She is a recipient of the Berkeley Fellowship, the NSF Graduate Fellowship, and the ACM Software System Award. Dr Hamrick’s work has been published in numerous venues including ICLR, ICML, NeurIPS, and PNAS.

The classical approach to AI was to design systems that were rational at run-time: they had explicit representations of beliefs, goals, and plans and ran inference algorithms, online, to select actions. The rational approach was criticised (by the behaviourists) and modified (by the probabilists) but persisted in some form. Now the overwhelming success of the connectionist approach in so many areas presents evidence that the rational view may no longer have a role to play in AI. This talk examines this question from several perspectives, including whether the rationality is present at design-time and/or at run-time, and whether systems with run-time rationality might be useful from the perspectives of computational efficiency, cognitive modelling and safety. It will present some current research focused on understanding the roles of learning in runtime-rational systems with the ultimate aim of constructing general-purpose human-level intelligent robots.

Professor Leslie Pack Kaelbling, Massachusetts Institute of Technology, USA

Professor Leslie Pack Kaelbling, Massachusetts Institute of Technology, USA

Leslie is a Professor at Massachusetts Institute of Technology. She has an undergraduate degree in Philosophy and a PhD in Computer Science from Stanford and was previously on the faculty at Brown University. She was the founding editor-in-chief of the Journal of Machine Learning Research. Her research agenda is to make intelligent robots using methods including estimation, learning, planning, and reasoning.

To flexibly adapt to new situations, our brains must understand the regularities in the world, but also in our own patterns of behaviour. A wealth of findings is beginning to reveal the algorithms we use to map the outside world. In contrast, the biological algorithms that map the complex structured behaviours we compose to reach our goals remain enigmatic. Here we reveal a neuronal implementation of an algorithm for mapping abstract behavioural structure and transferring it to new scenarios. We trained mice on many tasks which shared a common structure organising a sequence of goals but differed in the specific goal locations. Animals discovered the underlying task structure, enabling zero-shot inferences on the first trial of new tasks. The activity of most neurons in the medial Frontal cortex tiled progress-to-goal, akin to how place cells map physical space. These “goal-progress cells” generalised, stretching and compressing their tiling to accommodate different goal distances. In contrast, progress along the overall sequence of goals was not encoded explicitly. Instead, a subset of goal-progress cells was further tuned such that individual neurons fired with a fixed task-lag from a particular behavioural step. Together these cells implemented an algorithm that instantaneously encoded the entire sequence of future behavioural steps, and whose dynamics automatically retrieved the appropriate action at each step. These dynamics mirrored the abstract task structure both on-task and during offline sleep. Our findings suggest that goal-progress cells in the medial frontal cortex may be elemental building blocks of schemata that can be sculpted to represent complex behavioural structures.

Professor Timothy Behrens FRS, University of Oxford and University College London, UK

Professor Timothy Behrens FRS, University of Oxford and University College London, UK

Tim Behrens works at Oxford and University College London and is interested in how the front of the brain works.

Dr Fedorenko seeks to understand how humans understand and produce language, and how language relates to, and works together with, the rest of human cognition. She will discuss the ‘core’ language network, which includes left-hemisphere frontal and temporal areas, and show that this network is ubiquitously engaged during language processing across input and output modalities, strongly interconnected, and causally important for language. This language network plausibly stores language knowledge and supports linguistic computations related to accessing words and constructions from memory and combining them to interpret (decode) or generate (encode) linguistic messages. Importantly, the language network is sharply distinct from higher-level systems of knowledge and reasoning. First, the language areas show little neural activity when individuals solve math problems, infer patterns from data, or reason about others’ minds. And second, some individuals with severe aphasia lose the ability to understand and produce language but can still do math, play chess, and reason about the world. Thus, language does not appear to be necessary for thinking and reasoning. Human thinking instead relies on several brain systems, including the network that supports social reasoning and the network that supports abstract formal reasoning. These systems are sometimes engaged during language use—and thus have to work with the language system—but are not language-selective. Many exciting questions remain about the representations and computations in the systems of thought and about how the language system interacts with these higher-level systems. Furthermore, the sharp separation between language and thought in the human brain has implications for how we think about this relationship in the context of AI models, and for what we can expect from neural network models trained solely on linguistic input with the next-word prediction objective.

Professor Evelina Fedorenko, Massachusetts Institute of Technology, USA

Professor Evelina Fedorenko, Massachusetts Institute of Technology, USA

Dr Ev Fedorenko is a cognitive neuroscientist who studies the human language system. She received her bachelor’s degree from Harvard in 2002, and her PhD from Massachusetts Institute of Technology (MIT) in 2007. She was then awarded a K99R00 career development award from NIH. In 2014, she joined the faculty at MGH/HMS, and in 2019 she returned to MIT where she is currently an Associate Professor of Neuroscience in the BCS Department and the McGovern Institute for Brain Research. Dr Fedorenko uses fMRI, intracranial recordings and stimulation, EEG, MEG, and computational modelling, to study adults and children, including those with developmental and acquired brain disorders, and otherwise atypical brains.

Use of the symbol concept suffers from a conflation of two interpretations: a) a conventional sign vehicle (an alphanumeric character), and b) a conventional reference relationship (word meaning). Both are often mischaracterised in terms of "arbitrarity," a negative attribute. When your computer begins randomly displaying characters (a) on your screen, they are generally interpreted as indications of malfunction (or the operation of a "viral" algorithm). And yet when LLMs print out strings of characters that are iconic of interpretable sentences, we assume (b) that they are more than mere icons and indices of an algorithmic (aka mechanistic) process. This begs the question of what distinguishes symbolic interpretation from iconic and indexical interpretation and how they are related. Conventional relations are not just "given," however, they must be acquired. As a result, they are dependent on prior non-conventional referential relations (i.e. iconic and indexical interpretive processes) to extrinsically "ground" the reference of these intrinsically "ungrounded" sign vehicles. This semiotic infrastructure exemplifies the hierarchic complexity of symbolic reference, why it is cognitively difficult for non-humans, and hints at the special neurological architecture that aids human symbolic cognition. It also is relevant for understanding the difference between how humans and generative AI systems produce and process the tokens used as sign vehicles. So far, LLMs and their generative cousins are structured by token-token iconic and indexical relations only (though of extremely high dimensionality), not externally grounded by iconic and indexical pragmatic relations, even though the token-token relations of the training data have been.

Professor Terrence W Deacon, University of California, USA

Professor Terrence W Deacon, University of California, USA

Professor Deacon has held faculty positions at Harvard University, Harvard Medical School, Boston University, and the University of California, Berkeley, where he is currently the Davis Chair Distinguished Professor of Anthropology as well as on the faculty of the Institute for Cognitive and Brain Sciences. His laboratory research has focused on comparative and developmental neuroanatomy, particularly of humans, and includes the study of species differences using quantitative, physiological, and cross-species fetal neural transplantation techniques. His 1997 book The Symbolic Species: The Coevolution of Language and the Brain explored the evolution of the human brain and how it gave rise to our language abilities. His 2012 book Incomplete Nature: How Mind Emerged from Matter explored how interrelationships between thermodynamic, self-organising, semiotic, and evolutionary processes contributed to the emergence of life, mind, and human symbolic abilities.