Symbolic Behaviour in Artificial Intelligence

Santoro, A.; Lampinen, Andrew K.; Mathewson, Kory W.; Lillicrap, Timothy P.; Raposo, David

doi:10.48550/arxiv.2102.03406

Cited by 14 publications

(16 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Explanatory Learning enriches the fundamental problem of modern program synthesis (e.g., Balog et al, 2017;Ellis et al, 2020) by including the interpretation step among what should be learned. As seen with a few examples (see the Handling ambiguity paragraph), a CRN with learned I can exploit the ambiguity of language to impose new meaning on arbitrary substrates, which Santoro et al (2021) recognize as a fundamental trait of symbolic behavior. Recent literature finds few yet remarkable approaches that fit our EL paradigm, such as CLIP (Radford et al, 2021) in the vision area, and Generate & Rank (Shen et al, 2021) for Math Word Problems in NLP.…”

Section: Related Workmentioning

confidence: 99%

Explanatory Learning: Beyond Empiricism in Neural Networks

Norelli¹,

Mariani²,

Moschella³

et al. 2022

Preprint

View full text Add to dashboard Cite

We introduce Explanatory Learning (EL), a framework to let machines use existing knowledge buried in symbolic sequences -e.g. explanations written in hieroglyphic -by autonomously learning to interpret them. In EL, the burden of interpreting symbols is not left to humans or rigid human-coded compilers, as done in Program Synthesis. Rather, EL calls for a learned interpreter, built upon a limited collection of symbolic sequences paired with observations of several phenomena. This interpreter can be used to make predictions on a novel phenomenon given its explanation, and even to find that explanation using only a handful of observations, like human scientists do. We formulate the EL problem as a simple binary classification task, so that common end-to-end approaches aligned with the dominant empiricist view of machine learning could, in principle, solve it. To these models, we oppose Critical Rationalist Networks (CRNs), which instead embrace a rationalist view on the acquisition of knowledge. CRNs express several desired properties by construction, they are truly explainable, can adjust their processing at test-time for harder inferences, and can offer strong confidence guarantees on their predictions. As a final contribution, we introduce Odeen, a basic EL environment that simulates a small flatland-style universe full of phenomena to explain. Using Odeen as a testbed, we show how CRNs outperform empiricist end-to-end approaches of similar size and architecture (Transformers) in discovering explanations for novel phenomena. Explanation 1 : Explanation: Figure 1: The Odeen universe. A convenient environment to study and test the process of knowledge discovery in machines. Like the night sky was for humans. *Galileo did not sketch negative examples.

show abstract

Section: Related Workmentioning

confidence: 99%

Explanatory Learning: Beyond Empiricism in Neural Networks

Norelli¹,

Mariani²,

Moschella³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In short, language in decstr and imagine is more indexical than symbolic because the language tokens do not form a system Nieder (2009). The acquisition of symbolic behaviour is now an emerging topic Santoro et al (2021) which can be of fundamental importance for teachable autonomous agents as, from one side, considering a social partner is necessary to establish conventional meaning and, from the other side, such agents may need the flexibility of symbolic behaviour to appropriately learn from natural tutors.…”

Section: Open Questions and Prospects For Designing Agentsmentioning

confidence: 99%

Towards Teachable Autotelic Agents

Sigaud¹,

Akakzia²,

Caselles-Dupré³

et al. 2021

Preprint

View full text Add to dashboard Cite

Autonomous discovery and direct instruction are two extreme sources of learning in children, but educational sciences have shown that intermediate approaches such as assisted discovery or guided play resulted in better acquisition of skills. When turning to Artificial Intelligence, the above dichotomy can be translated into the distinction between autonomous agents, which learn in isolation from their own signals, and interactive learning agents which can be taught by social partners but generally lack autonomy. In between should stand teachable autonomous agents: agents that learn from both internal and teaching signals to benefit from the higher efficiency of assisted discovery processes. Designing such agents could result in progress in two ways. First, very concretely, it would offer a way to non-expert users in the real world to drive the learning behavior of agents towards their expectations. Second, more fundamentally, it might be a key step to endow agents with the necessary capabilities to reach general intelligence. The purpose of this paper is to elucidate the key obstacles standing in the way towards the design of such agents. We proceed in four steps. First, we build on a seminal work of Bruner to extract relevant features of the assisted discovery processes happening between a child and a tutor. Second, we highlight how current research on intrinsically motivated agents is paving the way towards teachable and autonomous agents. In particular, we focus on autotelic agents, i.e. agents equipped with forms of intrinsic motivations that enable them to represent, self-generate and pursue their own goals. We argue that such autotelic capabilities from the learner side are key in the discovery process. Third, we adopt a social learning perspective on the interaction between a tutor and a learner to highlight some components that are currently missing to these agents before they can be taught by ordinary people using natural pedagogy. Finally, we provide a list of specific research questions that emerge from the perspective of extending these agents with assisted learning capabilities.

show abstract

“…What is meaning, and how is it computed? Such questions are of broad practical significance even to the likes of Deepmind [11].…”

Section: Introductionmentioning

confidence: 99%

Intensional Artificial Intelligence: From Symbol Emergence to Explainable and Empathetic AI

Bennett¹,

Maruyama²

2021

Preprint

View full text Add to dashboard Cite

We argue that an explainable artificial intelligence must possess a rationale for its decisions, be able to infer the purpose of observed behaviour, and be able to explain its decisions in the context of what its audience understands and intends. To address these issues we present four novel contributions. Firstly, we define an arbitrary task in terms of perceptual states, and discuss two extremes of a domain of possible solutions. Secondly, we define the intensional solution. Optimal by some definitions of intelligence, it describes the purpose of a task. An agent possessed of it has a rationale for its decisions in terms of that purpose, expressed in a perceptual symbol system grounded in hardware. Thirdly, to communicate that rationale requires natural language, a means of encoding and decoding perceptual states. We propose a theory of meaning in which, to acquire language, an agent should model the world a language describes rather than the language itself. If the utterances of humans are of predictive value to the agent's goals, then the agent will imbue those utterances with meaning in terms of its own goals and perceptual states. In the context of Peircean semiotics, a community of agents must share rough approximations of signs, referents and interpretants in order to communicate. Meaning exists only in the context of intent, so to communicate with humans an agent must have comparable experiences and goals. An agent that learns intensional solutions, compelled by objective functions somewhat analogous to human motivators such as hunger and pain, may be capable of explaining its rationale not just in terms of its own intent, but in terms of what its audience understands and intends. It forms some approximation of the perceptual states of humans. Finally, we introduce the Mirror Symbol Hypothesis which posits that symbols learned as part of an intensional solution, by an agent labouring under human-like compulsions, behave like mirror neurons. These neurons, according to popular speculation, may play an important role in the facilitation of empathy.

show abstract

Symbolic Behaviour in Artificial Intelligence

Cited by 14 publications

References 53 publications

Explanatory Learning: Beyond Empiricism in Neural Networks

Explanatory Learning: Beyond Empiricism in Neural Networks

Towards Teachable Autotelic Agents

Intensional Artificial Intelligence: From Symbol Emergence to Explainable and Empathetic AI

Contact Info

Product

Resources

About