If you are looking for recordings of previous talks, visit this website instead.
You will need to login with a Mila account to access the recordings.
Predictive learning with apical dendrites in the hippocampus
Speaker: Dr. Colleen Gillon
Neural Flow Operators: Learning Dynamical Systems from Data
Speaker: Dr. Daniel Durstewitz
Curiosity Styles in the (Natural & Artificial) Wild
Speaker: Dr. Dani Bassett
Serotonin predictively encodes value
Human cortical cultures and artificial models: Understanding individuals with epilepsy
Principles of Network Plasticity in Neocortical Microcircuits
Speaker: Guillaume Durandau
Geometry of neural networks with biological structure: Randomness vs. structure and learning
Speaker: Kameron Decker Harris
Semantic scene descriptions as an objective of human vision
A theory of representation learning gives a deep generalisation of kernel methods, and implications for the brain
Speaker: Laurence Aitchison
Architectural foundations of flexible, robust and efficient cognitive motor control
Speaker: Laureline Logiaco
Network inference via process motifs for lagged correlation in linear stochastic processes
Representational Dissimilarity Metric Spaces for Stochastic Neural Networks
α-ReQ: Assessing Representation Quality in Self-Supervised Learning by measuring eigenspectrum decay
AI for neuroscience: data-driven models of information processing and learning
Speaker: Guillaume Bellec
Dopaminergic Reinforcement Learning Models for High-Dimensional Tasks
AI for brain interfaces: challenges and opportunities (Special talk organized with support from UNIQUE)
Modelling neural control of motion stabilization using recurrent neural networks
Biological reinforcement learning in hippocampus
Beyond accuracy: generalization properties of bio-plausible temporal credit assignment rules
Multi-Object Representation and Prediction in the Primate Brain
Speaker: Nicholas Watters
Characterizing Mental Representations using Deep Image Synthesis and Behavior
Speaker: Laurent Caplette
Neural representations of active vision
System 2 GFlowNets and System 2 Cognitive Neuroscience
When to retrieve and encode episodic memories: a neural network model of hippocampal-cortical interaction
On biological plausibility of weight sharing
Learning Hierarchical Temporal Representations through Dynamic Predictive Coding
Welcome Event, Winter 2022
Speakers: Blake Richards, Doina Precup, Guillaume Lajoie, Irina Rish, Karim Jerbi, Eilif Muller, Guillaume Dumas, Pouya Bashivan
Lifelong Learning: Theory and Practice
Speaker: Joshua T. Vogelstein
Structure, Function, and Learning in Distributed Neuronal Networks
Distributed TD Learning in a Heterogeneous Dopamine System
The story of a recurrent neural network’s training from initialization to task mastery
Visual Routines in brains and machines
A rapid and efficient learning rule for biological neural circuits
Speaker: Agnieszka Grabska-Barwińska
Vision as Bayesian inference: promise and pitfalls
Creative memory semantization through adversarial dreaming
Speaker: Nicolas Deperrois
The computational origin of human representation
Speaker: Steven Piantadosi
A unique coding of memories in the human brain
Speaker: Rodrigo Quian Quiroga
Context-gated Feature Representations for Multi-Task Learning in Humans and Neural Networks.
Impression learning: online predictive coding with synaptic plasticity
Speaker: Colin Bredenberg
What songbirds can tell us about reinforcement learning
Temporal credit assignment with eligibility traces and predictions
Speaker: Guillaume Bellec
A diversity of discounting horizons explains ramping diversity in dopamine-based reinforcement learning
Reward prediction errors shape memory during reinforcement learning
Inferring implicit inference (III)
Exploring the Cognitive-Motor Interface
Unsupervised deep learning identifies semantic disentanglement in single inferotemporal face patch neurons
The Temporal Dynamics of Opportunity Costs: A Normative Account of Cognitive Fatigue and Boredom
Replay slows down with experience
Speaker: Alice Berners-Lee
Behavioral algorithms: flexible, robust, and generalizable behavior in the natural world
Prioritized experience replay supports planning and non-local learning
Continual Reinforcement Learning with Memory at Multiple Timescales
Speaker: Christos Kaplanis
Computational mechanisms of selective attention during human reinforcement learning
Speaker: Angela Radulescu
Evidence for hierarchically-structured reinforcement learning in humans
The role of the hippocampus in memory and problem solving
Dynamic resource allocation during reinforcement learning accounts for ramping and phasic dopamine activity
Probabilistic successor representations allow for flexible behaviour
Prioritization of mental simulations for planning
Temporal and state abstractions for efficient learning, transfer and composition in humans
How, when and why do we make reflective versus reflexive choices
NeuroQuery: predictive meta-analysis of the neuroimaging literature
Towards New AI: Feeling Machines, Homeostatically Motivated Intelligence and Scale-free Cognition
A reductionist approach to interpreting neural networks that learn
Speaker: Maximilian Puelma Touzel
Half-Day Workshop on Neural Backpropagation
Speakers: Blake Richards, Irina Rish, Gaël Varoquaux, Yoshua Bengio
Inhibitory microcircuits for top-down plasticity of sensory representations
Speaker: Katharina Wilmes
Synergy between AI and Neuroscience
Dynamics of information transformation within a cortical circuit
Amplifying the Neural Power Spectrum
Continuation of Biological and Artificial RL
Neural population control via deep image synthesis
Reach out to us at neuralai.montreal@gmail.com if you have any questions!© 2023 The Mila Neural-AI Reading Group. The placeholder images were generated using Stable Diffusion.