Blake Richards

Biography

Blake Richards is Research Scientist Manager with the Paradigms of Intelligence team at Google, and an Associate Professor in the School of Computer Science and Department of Neurology and Neurosurgery at McGill University. He is also a Core Faculty Member at Mila.

Richards’ research lies at the intersection of neuroscience and AI. His laboratory investigates universal principles of intelligence that apply to both natural and artificial agents.

He has received several awards for his work, including the NSERC Arthur B. McDonald Fellowship in 2022, the Canadian Association for Neuroscience Young Investigator Award in 2019, and a Canada CIFAR AI Chair in 2018. Richards was a Banting Postdoctoral Fellow at SickKids Hospital from 2011 to 2013.

He obtained his PhD in neuroscience from the University of Oxford in 2010, and his BSc in cognitive science and AI from the University of Toronto in 2004.

Current Students

Aliya Affdal

Collaborating researcher - Université de Montréal

aliya.affdal@mila.quebec

Benjamin Alsbury-Nealy

PhD - McGill University

benjamin@silicolabs.ca

antoine.boudreau-leblanc@mila.quebec

Antoine Boudreau LeBlanc

Postdoctorate - McGill University

Colin Bredenberg

Postdoctorate - Université de Montréal

Principal supervisor :

colin.bredenberg@mila.quebec

Nour Chahine

PhD - McGill University

nour.chahine@mila.quebec

erica.cianfarano@mila.quebec

Raymond Chua

PhD - McGill University

Principal supervisor :

PhD - McGill University

Jonathan Cornford

Collaborating researcher - McGill University

cornforj@mila.quebec

Sabrina Du

Undergraduate - McGill University

sabrina.du@mila.quebec

aleksei.efremov@mail.mcgill.ca

Alex Efremov

PhD - McGill University

Tom George

Postdoctorate - McGill University

Co-supervisor :

tomgeorge1@btinternet.com

Independent visiting researcher - Université de Montréal

oumayma.gharbi@mila.quebec

Arna Ghosh

Collaborating researcher - McGill University

Adel Halawa

PhD - McGill University

adel.halawa@mila.quebec

Roy Henha Eyono

PhD - McGill University

roy.eyono@mila.quebec

Sonia Joseph

PhD - McGill University

sonia.joseph@mila.quebec

daniel.levenstein@mila.quebec

Daniel Levenstein

Collaborating Alumni - McGill University

Pingsheng Li

PhD - McGill University

Co-supervisor :

pingsheng.li@mila.quebec

matthew.loukine@mail.mcgill.ca

Matthew Loukine

PhD - McGill University

Dane Malenfant

Master's Research - McGill University

dane.malenfant@mila.quebec

Abdel Mfougouon Njupoun

PhD - Université de Montréal

Principal supervisor :

le-thuy-duong.nguyen@mila.quebec

abdelnjupoun@gmail.com

Master's Research - McGill University

Principal supervisor :

Matt Perich

qazaleh.ranjbaran@gmail.com

Ghazaleh Ranjbaran

Independent visiting researcher - Université de Montréal

Alexis Roger

PhD - McGill University

Co-supervisor :

Irina Rish

alexis.roger@mila.quebec

ali.sahebpasand@mila.quebec

Ali Saheb Pasand

PhD - McGill University

Co-supervisor :

Pablo Samuel Castro

Mandana Samiei

PhD - McGill University

Principal supervisor :

Doina Precup

samiemandana@gmail.com

Aidan Sirbu

Master's Research - McGill University

aidan.sirbu@mila.quebec

hiroyuki.tanabe@mila.quebec

Hiro Tanabe

Independent visiting researcher - NA

Josh Tindall

Collaborating Alumni - McGill University

joshua.tindall@mila.quebec

Mashbayar Tugsbayar

PhD - McGill University

charlotte.volk@mila.quebec

Charlotte Volk

PhD - McGill University

Co-supervisor :

Shahab Bakhtiari

maren.wehrheim@mila.quebec

Maren Wehrheim

Independent visiting researcher - York University

amirhossein.zamani@mila.quebec

Amirhossein Zamani

PhD - Concordia University

Principal supervisor :

Eugene Belilovsky

Machine Learning for the Segmentation of Different Nerve Fibre Activations from Brain-to-body Neural Signals

Blog Posts

Représentation graphique d'un nerf vague

May 21, 2025

Param Raval

Olivier Tessier-Larivière

Pascal Fortier-Poisson

Blake Richards

Guillaume Lajoie

Read the article

June 13, 2024

What Do Synaptic Weight Distributions Tell Us About Learning in the Brain ?

Roman Pogodin

Jonathan Cornford

Arna Ghosh

Gauthier Gidel

Guillaume Lajoie

Blake Richards

Read the article

August 29, 2023

α-ReQ: Assessing Representation Quality in SSL

KK Agrawal

Arnab Kumar-Mondal

Arna Ghosh

Blake A. Richards

Read the article

Publications

Sequential predictive learning is a unifying theory for hippocampal representation and replay

Aleksei Efremov

The mammalian hippocampus contains a cognitive map that represents an animal’s position in the environment 1 … (see more) and generates offline “replay” 2,3 for the purposes of recall 4 , planning 5,6 , and forming long term memories 7 . Recently, it’s been found that artificial neural networks trained to predict sensory inputs develop spatially tuned cells 8 , aligning with predictive theories of hippocampal function 9–11 . However, whether predictive learning can also account for the ability to produce offline replay is unknown. Here, we find that spatially-tuned cells, which robustly emerge from all forms of predictive learning, do not guarantee the presence of a cognitive map with the ability to generate replay. Offline simulations only emerged in networks that used recurrent connections and head-direction information to predict multi-step observation sequences, which promoted the formation of a continuous attractor reflecting the geometry of the environment. These offline trajectories were able to show wake-like statistics, autonomously replay recently experienced locations, and could be directed by a virtual head direction signal. Further, we found that networks trained to make cyclical predictions of future observation sequences were able to rapidly learn a cognitive map and produced sweeping representations of future positions reminiscent of hippocampal theta sweeps 12 . These results demonstrate how hippocampal-like representation and replay can emerge in neural networks engaged in predictive learning, and suggest that hippocampal theta sequences reflect a circuit that implements a data-efficient algorithm for sequential predictive learning. Together, this framework provides a unifying theory for hippocampal functions and hippocampal-inspired approaches to artificial intelligence.

2024-06-03

bioRxiv (preprint)

Fast burst fraction transients convey information independent of the firing rate

Richard Naud

Xingyun Wang

Zachary Friedenberger

Alexandre Payeur

Jiyun N Shin

Jean-Claude Beique

Moritz Drüke

Matthew Larkum

Guy Doron

Theories of attention and learning have hypothesized a central role for high-frequency bursting in cognitive functions, but experimental rep… (see more)orts of burst-mediated representations \emph{in vivo} have been limited. Here we used a novel demultiplexing approach by considering a conjunctive burst code. We studied this code \emph{in vivo} while animals learned to report direct electrical stimulation of the somatosensory cortex and found two acquired yet independent representations. One code, the event rate, showed a sparse and succint stiumulus representation and a small modulation upon detection errors. The other code, the burst fraction, correlated more globally with stimulation and more promptly responded to detection errors. Potent and fast modulations of the burst fraction were seen even in cells that were considered unresponsive based on the firing rate. During the later stages of training, this modulation in bursting happened earlier, gradually aligning temporally with the representation in event rate. The alignment of bursting and event rate modulation sharpened the firing rate response, and was strongly associated with behavioral accuracy. Thus a fine-grained separation of spike timing patterns reveals two signals that accompany stimulus representations: an error signal that can be essential to guide learning and a sharpening signal that could implement attention mechanisms.

2024-03-26

bioRxiv (preprint)

A Unified, Scalable Framework for Neural Population Decoding

Mehdi Azabou

Vinam Arora

Venkataramana Ganesh

Ximeng Mao

Santosh Nachimuthu

Michael J. Mendelson

Matthew G. Perich

Eva L. Dyer

Our ability to use deep learning approaches to decipher neural activity would likely benefit from greater scale, in terms of both model size… (see more) and datasets. However, the integration of many neural recordings into one unified model is challenging, as each recording contains the activity of different neurons from different individual animals. In this paper, we introduce a training framework and architecture designed to model the population dynamics of neural activity across diverse, large-scale neural recordings. Our method first tokenizes individual spikes within the dataset to build an efficient representation of neural events that captures the fine temporal structure of neural activity. We then employ cross-attention and a PerceiverIO backbone to further construct a latent tokenization of neural population activities. Utilizing this architecture and training framework, we construct a large-scale multi-session model trained on large datasets from seven nonhuman primates, spanning over 158 different sessions of recording from over 27,373 neural units and over 100 hours of recordings. In a number of different tasks, we demonstrate that our pretrained model can be rapidly adapted to new, unseen sessions with unspecified neuron correspondence, enabling few-shot performance with minimal labels. This work presents a powerful new approach for building deep learning tools to analyze neural data and stakes out a clear path to training at scale.

2023-12-11

Neural Information Processing Systems (Accept (poster))

The feature landscape of visual cortex

Rudi Tong

Ronan da Silva

James Wilsenach

Stuart Trenholm

Understanding computations in the visual system requires a characterization of the distinct feature preferences of neurons in different visu… (see more)al cortical areas. However, we know little about how feature preferences of neurons within a given area relate to that area’s role within the global organization of visual cortex. To address this, we recorded from thousands of neurons across six visual cortical areas in mouse and leveraged generative AI methods combined with closed-loop neuronal recordings to identify each neuron’s visual feature preference. First, we discovered that the mouse’s visual system is globally organized to encode features in a manner invariant to the types of image transformations induced by self-motion. Second, we found differences in the visual feature preferences of each area and that these differences generalized across animals. Finally, we observed that a given area’s collection of preferred stimuli (‘own-stimuli’) drive neurons from the same area more effectively through their dynamic range compared to preferred stimuli from other areas (‘other-stimuli’). As a result, feature preferences of neurons within an area are organized to maximally encode differences among own-stimuli while remaining insensitive to differences among other-stimuli. These results reveal how visual areas work together to efficiently encode information about the external world.

2023-11-04

bioRxiv (preprint)

Contrastive Retrospection: honing in on critical steps for rapid learning and generalization in RL

Chen Sun

Wannan Yang

Thomas Jiralerspong

Dane Malenfant

Benjamin Alsbury-Nealy

Yoshua Bengio

In real life, success is often contingent upon multiple critical steps that are distant in time from each other and from the final reward. T… (see more)hese critical steps are challenging to identify with traditional reinforcement learning (RL) methods that rely on the Bellman equation for credit assignment. Here, we present a new RL algorithm that uses offline contrastive learning to hone in on these critical steps. This algorithm, which we call Contrastive Retrospection (ConSpec), can be added to any existing RL algorithm. ConSpec learns a set of prototypes for the critical steps in a task by a novel contrastive loss and delivers an intrinsic reward when the current state matches one of the prototypes. The prototypes in ConSpec provide two key benefits for credit assignment: (i) They enable rapid identification of all the critical steps. (ii) They do so in a readily interpretable manner, enabling out-of-distribution generalization when sensory features are altered. Distinct from other contemporary RL approaches to credit assignment, ConSpec takes advantage of the fact that it is easier to retrospectively identify the small set of steps that success is contingent upon (and ignoring other states) than it is to prospectively predict reward at every taken step. ConSpec greatly improves learning in a diverse set of RL tasks. The code is available at the link: https://github.com/sunchipsster1/ConSpec

2023-09-20

Neural Information Processing Systems (poster)

Learning better with Dale's Law: A Spectral Perspective

Most recurrent neural networks (RNNs) do not include a fundamental constraint of real neural circuits: Dale’s Law, which implies that neur… (see more)ons must be excitatory (E) or inhibitory (I). Dale’s Law is generally absent from RNNs because simply partitioning a standard network’s units into E and I populations impairs learning. However, here we extend a recent feedforward bio-inspired EI network architecture, named Dale’s ANNs, to recurrent networks, and demonstrate that good performance is possible while respecting Dale’s Law. This begs the question: What makes some forms of EI network learn poorly and others learn well? And, why does the simple approach of incorporating Dale’s Law impair learning? Historically the answer was thought to be the sign constraints on EI network parameters, and this was a motivation behind Dale’s ANNs. However, here we show the spectral properties of the recurrent weight matrix at initialisation are more impactful on network performance than sign constraints. We find that simple EI partitioning results in a singular value distribution that is multimodal and dispersed, whereas standard RNNs have an unimodal, more clustered singular value distribution, as do recurrent Dale’s ANNs. We also show that the spectral properties and performance of partitioned EI networks are worse for small networks with fewer I units, and we present normalised SVD entropy as a measure of spectrum pathology that correlates with performance. Overall, this work sheds light on a long-standing mystery in neuroscience-inspired AI and computational neuroscience, paving the way for greater alignment between neural networks and biology.

2023-09-20

NeurIPS.cc/2023/Conference (poster)

Deep neural networks are not a single hypothesis but a language for expressing computational hypotheses

Tal Golan

JohnMark Taylor

Heiko Herbert Schütt

Benjamin Peters

Rowan Paolo Sommers

Katja Seeliger

Adrien Doerig

Paul Linton

Talia Konkle

Marcel van Gerven

Konrad Kording

Tim Christian Kietzmann

Grace W. Lindsay

Nikolaus Kriegeskorte

An ideal vision model accounts for behavior and neurophysiology in both naturalistic conditions and designed lab experiments. Unlike psyc… (see more)hological theories, artificial neural networks (ANNs) actually perform visual tasks and generate testable predictions for arbitrary inputs. These advantages enable ANNs to engage the entire spectrum of the evidence. Failures of particular models drive progress in a vibrant ANN research program of human vision.

2022-12-31

Behavioral and Brain Sciences (published)

Formalizing locality for normative synaptic plasticity models

Cristina Savin

2022-12-31

Advances in Neural Information Processing Systems (published)

The application of artificial intelligence to biology and neuroscience

Doris Tsao

Anthony Zador

2022-06-30

Cell (published)

Towards Scaling Difference Target Propagation by Learning Backprop Targets

The development of biologically-plausible learning algorithms is important for understanding learning in the brain, but most of them fail to… (see more) scale-up to real-world tasks, limiting their potential as explanations for learning by real brains. As such, it is important to explore learning algorithms that come with strong theoretical guarantees and can match the performance of backpropagation (BP) on complex tasks. One such algorithm is Difference Target Propagation (DTP), a biologically-plausible learning algorithm whose close relation with Gauss-Newton (GN) optimization has been recently established. However, the conditions under which this connection rigorously holds preclude layer-wise training of the feedback pathway synaptic weights (which is more biologically plausible). Moreover, good alignment between DTP weight updates and loss gradients is only loosely guaranteed and under very specific conditions for the architecture being trained. In this paper, we propose a novel feedback weight training scheme that ensures both that DTP approximates BP and that layer-wise feedback weight training can be restored without sacrificing any theoretical guarantees. Our theory is corroborated by experimental results and we report the best performance ever achieved by DTP on CIFAR-10 and ImageNet 32

2022-06-27

Proceedings of the 39th International Conference on Machine Learning (published)

proceedings.mlr.press

On Neural Architecture Inductive Biases for Relational Tasks

Current deep learning approaches have shown good in-distribution generalization performance, but struggle with out-of-distribution generaliz… (see more)ation. This is especially true in the case of tasks involving abstract relations like recognizing rules in sequences, as we find in many intelligence tests. Recent work has explored how forcing relational representations to remain distinct from sensory representations, as it seems to be the case in the brain, can help artificial systems. Building on this work, we further explore and formalize the advantages afforded by 'partitioned' representations of relations and sensory details, and how this inductive bias can help recompose learned relational structure in newly encountered settings. We introduce a simple architecture based on similarity scores which we name Compositional Relational Network (CoRelNet). Using this model, we investigate a series of inductive biases that ensure abstract relations are learned and represented distinctly from sensory data, and explore their effects on out-of-distribution generalization for a series of relational psychophysics tasks. We find that simple architectural choices can outperform existing models in out-of-distribution generalization. Together, these results show that partitioning relational representations from other information streams may be a simple way to augment existing network architectures' robustness when performing out-of-distribution relational computations.

2022-06-08

ArXiv (preprint)

arxiv.org

Time cell encoding is decoupled from time perception in deep reinforcement learning agents

Ann Zixiang Huang

Dongyan Lin

2021-12-31

Conference on Cognitive Computational Neuroscience (published)