AIConfPaper
← Search

Adam Santoro

21 accepted papers

2025

Tracing the Representation Geometry of Language Models from Pretraining to Post-training

Melody Zixuan Li, Kumar Krishna Agrawal, Arna Ghosh, Komal Kumar Teru, Adam Santoro, Guillaume Lajoie +1
NeurIPS 2025poster

Standard training metrics like loss fail to explain the emergence of complex capabilities in large language models. We take a spectral approach to investigate the geometry of learned representations across pretraining and post-training, measuring effective rank (RankMe) and eigenspectrum decay (αReQ…

Cited by 0Source ↗Scholar ↗
2022

A data-driven approach for learning to control computers

Peter C Humphreys, David Raposo, Tobias Pohlen, Gregory Thornton, Rachita Chhaparia, Alistair Muldal +4
ICML 2022spotlight

It would be useful for machines to use computers as humans do so that they can aid us in everyday tasks. This is a setting in which there is also the potential to leverage large-scale expert demonstrations and human judgements of interactive behaviour, which are two ingredients that have driven much…

Cited by 119Source ↗PDF ↗Scholar ↗
2022

Data Distributional Properties Drive Emergent In-Context Learning in Transformers

Stephanie C.Y. Chan, Adam Santoro, Andrew Kyle Lampinen, Jane X Wang, Aaditya K Singh, Pierre Harvey Richemond +2
NeurIPS 2022accept

Large transformer-based models are able to perform in-context few-shot learning, without being explicitly trained for it. This observation raises the question: what aspects of the training regime lead to this emergent behavior? Here, we show that this behavior is driven by the distributions of the t…

Cited by 335Source ↗PDF ↗code ↗Scholar ↗
2022

Intra-agent speech permits zero-shot task acquisition

Chen Yan, Federico Carnevale, Petko Georgiev, Adam Santoro, Aurelia Guy, Alistair Muldal +4
NeurIPS 2022accept

Human language learners are exposed to a trickle of informative, context-sensitive language, but a flood of raw sensory data. Through both social language use and internal processes of rehearsal and practice, language learners are able to build high-level, semantic representations that explain their…

Cited by 11Source ↗PDF ↗Scholar ↗
2022

Tell me why! Explanations support learning relational and causal structure

Andrew K Lampinen, Nicholas Roy, Ishita Dasgupta, Stephanie Cy Chan, Allison Tam, James Mcclelland +5
ICML 2022spotlight

Inferring the abstract relational and causal structure of the world is a major challenge for reinforcement-learning (RL) agents. For humans, language{—}particularly in the form of explanations{—}plays a considerable role in overcoming this challenge. Here, we show that language can play a similar ro…

Cited by 46Source ↗PDF ↗code ↗Scholar ↗
2021

Attention over Learned Object Embeddings Enables Complex Visual Reasoning

David Ding, Felix Hill, Adam Santoro, Malcolm Reynolds, Matthew Botvinick
NeurIPS 2021oral

Neural networks have achieved success in a wide array of perceptual tasks but often fail at tasks involving both perception and higher-level reasoning. On these more challenging tasks, bespoke approaches (such as modular symbolic components, independent dynamics models or semantic parsers) targeted…

Cited by 80Source ↗PDF ↗code ↗Scholar ↗
2021

Rapid Task-Solving in Novel Environments

Samuel Ritter, Ryan Faulkner, Laurent Sartran, Adam Santoro, Matthew Botvinick, David Raposo
ICLR 2021poster

We propose the challenge of rapid task-solving in novel environments (RTS), wherein an agent must solve a series of tasks as rapidly as possible in an unfamiliar environment. An effective RTS agent must balance between exploring the unfamiliar environment and solving its current task, all while buil…

Cited by 34Source ↗PDF ↗Scholar ↗
2020

Automated curriculum generation through setter-solver interactions

Sebastien Racaniere, Andrew Lampinen, Adam Santoro, David Reichert, Vlad Firoiu, Timothy Lillicrap
ICLR 2020poster

Reinforcement learning algorithms use correlations between policies and rewards to improve agent performance. But in dynamic or sparsely rewarding environments these correlations are often too small, or rewarding events are too infrequent to make learning feasible. Human education instead relies…

Cited by 44Source ↗Scholar ↗
2020

Environmental drivers of systematicity and generalization in a situated agent

Felix Hill, Andrew Lampinen, Rosalia Schneider, Stephen Clark, Matthew Botvinick, James L. McClelland +1
ICLR 2020poster

The question of whether deep neural networks are good at generalising beyond their immediate training experience is of critical importance for learning-based approaches to AI. Here, we consider tests of out-of-sample generalisation that require an agent to respond to never-seen-before instructions b…

Cited by 112Source ↗Scholar ↗
2019

An Investigation of Model-Free Planning

Arthur Guez, Mehdi Mirza, Karol Gregor, Rishabh Kabra, Sebastien Racaniere, Theophane Weber +7
ICML 2019oral

The field of reinforcement learning (RL) is facing increasingly challenging domains with combinatorial complexity. For an RL agent to address these challenges, it is essential that it can plan effectively. Prior work has typically utilized an explicit model of the environment, combined with a specif…

Cited by 108Source ↗PDF ↗code ↗Scholar ↗
2019

Deep reinforcement learning with relational inductive biases

Vinicius Zambaldi, David Raposo, Adam Santoro, Victor Bapst, Yujia Li, Igor Babuschkin +10
ICLR 2019poster

We introduce an approach for augmenting model-free deep reinforcement learning agents with a mechanism for relational reasoning over structured representations, which improves performance, learning efficiency, generalization, and interpretability. Our architecture encodes an image as a set of vector…

Cited by 265Source ↗PDF ↗Scholar ↗
2019

Hyperbolic Attention Networks

Caglar Gulcehre, Misha Denil, Mateusz Malinowski, Ali Razavi, Razvan Pascanu, Karl Moritz Hermann +5
ICLR 2019poster

Recent approaches have successfully demonstrated the benefits of learning the parameters of shallow networks in hyperbolic space. We extend this line of work by imposing hyperbolic geometry on the embeddings used to compute the ubiquitous attention mechanisms for different neural networks architectu…

Cited by 302Source ↗PDF ↗Scholar ↗
2019

Learning to Make Analogies by Contrasting Abstract Relational Structure

Felix Hill, Adam Santoro, David Barrett, Ari Morcos, Timothy Lillicrap
ICLR 2019poster

Analogical reasoning has been a principal focus of various waves of AI research. Analogy is particularly challenging for machines because it requires relational structures to be represented such that they can be flexibly applied across diverse domains of experience. Here, we study how analogical rea…

Cited by 102Source ↗PDF ↗code ↗Scholar ↗
2018

Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures

Sergey Bartunov, Adam Santoro, Blake Richards, Luke Marris, Geoffrey E. Hinton, Timothy Lillicrap
NeurIPS 2018poster

The backpropagation of error algorithm (BP) is impossible to implement in a real brain. The recent success of deep networks in machine learning and AI, however, has inspired proposals for understanding how the brain might learn across multiple layers, and hence how it might approximate BP. As of yet…

Cited by 333Source ↗PDF ↗code ↗Scholar ↗
2018

Learning Visual Question Answering by Bootstrapping Hard Attention

Mateusz Malinowski, Carl Doersch, Adam Santoro, Peter Battaglia
ECCV 2018poster

Attention mechanisms in biological perception are thought to select subsets of perceptual information for more sophisticated processing which would be prohibitive to perform on all sensory inputs. In computer vision, however, there has been relatively little exploration of hard attention, where some…

Cited by 132Source ↗PDF ↗Scholar ↗
2018

Measuring abstract reasoning in neural networks

David Barrett, Felix Hill, Adam Santoro, Ari Morcos, Timothy Lillicrap
ICML 2018oral

Whether neural networks can learn abstract reasoning or whether they merely rely on superficial statistics is a topic of recent debate. Here, we propose a dataset and challenge designed to probe abstract reasoning, inspired by a well-known human IQ test. To succeed at this challenge, models must cop…

Cited by 283Source ↗PDF ↗code ↗Scholar ↗
2018

Relational recurrent neural networks

Adam Santoro, Ryan Faulkner, David Raposo, Jack Rae, Mike Chrzanowski, Theophane Weber +4
NeurIPS 2018poster

Memory-based neural networks model temporal data by leveraging an ability to remember information for long periods. It is unclear, however, whether they also have an ability to perform complex relational reasoning with the information they remember. Here, we first confirm our intuitions that standar…

Cited by 283Source ↗PDF ↗code ↗Scholar ↗
2017

A simple neural network module for relational reasoning

Adam Santoro, David Raposo, David G Barrett, Mateusz Malinowski, Razvan Pascanu, Peter Battaglia +1
NeurIPS 2017spotlight

Relational reasoning is a central component of generally intelligent behavior, but has proven difficult for neural networks to learn. In this paper we describe how to use Relation Networks (RNs) as a simple plug-and-play module to solve problems that fundamentally hinge on relational reasoning. We t…

Cited by 2023Source ↗PDF ↗Scholar ↗
2017

Cognitive Psychology for Deep Neural Networks: A Shape Bias Case Study

Samuel Ritter, David G. T. Barrett, Adam Santoro, Matt M. Botvinick
ICML 2017poster

Deep neural networks (DNNs) have advanced performance on a wide range of complex tasks, rapidly outpacing our understanding of the nature of their solutions. While past work sought to advance our understanding of these models, none has made use of the rich history of problem descriptions, theories,…

Cited by 260Source ↗PDF ↗Scholar ↗
2017

Discovering objects and their relations from entangled scene representations

David Raposo, Adam Santoro, David Barrett, Razvan Pascanu, Timothy Lillicrap, Peter Battaglia
ICLR 2017workshop

Our world can be succinctly and compactly described as structured scenes of objects and relations. A typical room, for example, contains salient objects such as tables, chairs and books, and these objects typically relate to each other by virtue of their correlated features, such as position, functi…

Cited by 133Source ↗Scholar ↗
2016

Meta-Learning with Memory-Augmented Neural Networks

Adam Santoro, Sergey Bartunov, Matthew Botvinick, Daan Wierstra, Timothy Lillicrap
ICML 2016poster

Despite recent breakthroughs in the applications of deep neural networks, one setting that presents a persistent challenge is that of "one-shot learning." Traditional gradient-based networks require a lot of data to learn, often through extensive iterative training. When new data is encountered, the…

Cited by 3349Source ↗PDF ↗Scholar ↗