01/

Konrad Kording

• Experimental psychologists concerns regarding localization of brain function 

• Epistemic issues about decomposition of function

• Focused critique of neuroscience methods

• Ability of standard methods to decode even a "neuro" algorithm

• Grow neurons outside of the brain and study their independent behaviour

• Computing by time

02/

Adrienne Fairhall

• None listed

03/

Melanie Mitchell

• An Evolutionary Perspective on Biological and ANNs

• On the Measure of Intelligence

• Relationships in Analogical Reasoning

• Do Better ImageNet Models Transfer Better?

• Continual Learning

04/

Anthony Zador

• More is Different

05/

Marcin Miłkowski

• Short overview of Fodor's ideas

• Modularity of Mind

• Affordances in contemporary reinforcement learning

• Why normative models in neuroscience

• Integrated Information Theory

• Super Turing Machines 1

• Super Turing Machines 2

• Origin of Common Sense

• Anomalous Monism

• Neuromorphic Computing

• Building Machines That Learn And Think Like People

• Pancomputationalism

• Meta Meta

• Extended spider cognition

• Environment's role in cognition

• Locating the engram

• Homeostatic reinforcement learning

• Collective Memories 1

• Collective Memories 2

06/

Paul Cisek

• Materialism Eliminative

• Principles of Brain Evolution

• Evolutionary Neuroscience

• A Theory of Affordances in Reinforcement Learning

• Précis of principles of brain evolution

• A Modular Approach to Vocal Learning

• The Theory of Affordances

• The emergence of grammar

• Homeostatic reinforcement learning

• Paul Cisek about the semantic mess of "representations"

• Animal Cultures

07/

Recap and Panel Discussion

• Complexity & Stupidity

• What is Computation?

• No one knows what attention is

08/

John Langford

• Of exactitude in science

• Somatic_hypermutation

• Placing language in an integrated understanding system

• Humpback whale song hierarchical structure

• Syntactic features in humpback whale song

09/

Blake Richards

• Hippocampal Contributions to Control

• Optimizing agent behavior over long time scales by transporting value

• Marc Howard

• Hidden neural states underlie canary song syntax

• The Mind of a Mnemonist

• Complementary learning systems within the hippocampus

• Consistent Nonparametric Regression

• On the Strong Universal Consistency of Nearest Neighbor Regression

• The primacy of behavioral research for understanding the brain

• Tensor Product Variable Binding and the Representation of Symbolic Structures in Connectionist Systems

• On the proper treatment of connectionism

• Sparse Distributed Models

• What Constitutes an Episode in Episodic Memory?

• Reward prediction errors create event boundaries in memory

• A Scale-Invariant Internal Representation of Time

• Learning Structures: Predictive Representations, Replay, and Generalization

• Inhibitory engrams in perception and memory

• Janice Chen

10/

Jay McClelland

• Environmental drivers of systematicity and generalization in a situated agent

• Placing language in an integrated understanding system

• Aetherial Symbols

• Prefrontal cortex as a meta-reinforcement learning system

• Cultural diversity in chimpanzees

• Formal Statement

• Are Neural Nets Modular?

• Compositional Deep Learning

• Measuring Compositionality in Representation Learning

• Factorizing Declarative and Procedural Knowledge in Dynamical Systems

• Learning Basic Visual Concepts with a Constrained Variational Framework

• The Scattering Compositional Learner

11/

Gary Marcus

• Honey bees can perform delay-match-to-sample

• Compositional Reasoning in Early Childhood

• Addition and subtraction by human infants

• A unified account of numerosity perception

• Optogenetic Editing Reveals the Hierarchical Organization of Learned Action Sequences

• Hand Knob Area of Premotor Cortex Represents the Whole Body in a Compositional Way

• Journal of compositionality

• DNN or k-NN

• A brief survey of classic work on compositionality

• Compositionality

• Formal Semantics Pragmatics: Origins, Issues, Impact

• Compositionality as an Empirical Problem

• Performance vs. competence in human–machine comparisons

• Mental Structures

• Object Files and Schemata: Factorizing Declarative and Procedural Knowledge in Dynamical Systems

• Learning Physical Graph Representations from Visual Scenes

• On Meaning, Form, and Understanding in the Age of Data

• Object Representations for Learning and Reasoning

• The inverted T-model

• Entity Abstraction in Visual Model-Based Reinforcement Learning

• Improving Generative Imagination in Object-Centric World Models

• Compositionality as an empirical problem

12/

Jessica Hamrick

• Structured agents for physical construction

• The Spatial Memory Pipeline

• Generalization as Search

• When to use parametric models in reinforcement learning?

• Model-based RL in Contextual Decision Processes

• State, Transition and Learning Interactions in the Two-Step Task

• Predictive representations can link modelbased reinforcement learning to model-free mechanisms

• Is a Good Representation Sufficient for Sample Efficient Reinforcement Learning?

• Cerebellar Involvement in Anticipating the Consequences of Self-Produced Actions During Bimanual Movements

• Acme: a research framework for reinforcement learning

• Monte-Carlo Tree Search (MCTS)

• Tutorial on Model-Based Methods in Reinforcement Learning

• The successor representation in human reinforcement learning

• Human-level concept learning through probabilistic program induction

• ALFWorld: Aligning Text and Embodied Environments for Interactive Learning

• Planning to Explore via Self-Supervised World Models

• Investigating Human Priors for Playing Video Games

• Blueberry Earth

• On the role of planning in model-based deep reinforcement learning

• A normative account of confirmatory biases during reinforcement learning

13/

Mazviita Chirimuuta

• Outside Color

• The Algorithmic Lens

• Is coding a relevant metaphor for the brain?

• Do you compute?

• The Church-Turing Thesis

• The Halting Problem

• Representation of Events in Nerve Nets and Finite Automata

14/

Iris van Rooij

• Why Philosophers Should Care About Computational Complexity

• Levels of biological plausibility

• Satisficing

• Similar network activity from disparate circuit parameters

• Every Model Learned by Gradient Descent Is Approximately a Kernel Machine

• Monoids, weighted automata and algorithmic philosophy of science

• Against Method

15/

Recap and Panel Discussion

• Virtual Neuroscience with The Learning Salon

• Can a bird brain do phonology?

• Baffling: a condition-dependent alternative mate attraction strategy using self-made tools in tree crickets

• Tree crickets optimize the acoustics of baffles to exaggerate their mate-attraction signal

• The Model-less Neuromimetic Chip and its Normalization of Neuroscience and Artificial Intelligence

• The Foundations of Computing

• The maps problem and the mapping problem: Two challenges for a cognitive neuroscience of speech and language

• Probably Approximately Correct: Nature's Algorithms for Learning and Prospering in a Complex World

• The Contribution of Area MT to Visual Motion Perception Depends on Training

• Acute off-target effects of neural circuit manipulations

• Reverse hierarchies and sensory learning

• Object Files and Schemata: Factorizing Declarative and Procedural Knowledge in Dynamical Systems

• ON THE RELATIONSHIP OF ACTIVE INFERENCE AND CONTROL AS INFERENCE

• On the Origin of Objects

• And the New Yorker Cartoon Contest Winner Is … a Computer

• Causal Inference Book

• Intrinsic Social Motivation via Causal Influence in Multi-Agent RL

• The Animal-AI Testbed

• Scaling Laws for Neural Language Models

• Memory and the Computational Brain

16/

Jane Wang

• Jane's website

• Meta-learning in natural and artificial intelligence

• Matt Botvinick: Neuroscience, Psychology, and AI at DeepMind

• Meta-Learning in Neural Networks: A Survey

• Leslie Valiant: Evolution as Learning

• Biological evolution as a form of learning

• Probably Approximately Correct: Nature's Algorithms for Learning and Prospering in a Complex World

• Dopaminergic neurons write and update memories with cell-type-specific rules

• Meta-learning of Sequential Strategies

• Meta-trained agents implement Bayes-optimal agents

• The learning curve: Implications of a quantitative analysis

• Statistical Decision Theory and Bayesian Analysis

• CAUSALWORLD: A ROBOTIC MANIPULATION BENCHMARK FOR CAUSAL STRUCTURE AND TRANSFER LEARNING

• Mathematical Statistics: Basic Ideas And Selected Topics

• Task set and prefrontal cortex

• Prefrontal interactions reflect future task operations

• Resolving the paradox of evolvability with learning theory: How evolution learns to improve evolvability on rugged fitness landscapes

• Serotonin neurons modulate learning rate through uncertainty

• Transforming task representations to perform novel tasks

• Gödel machine

• Jürgen Schmidhuber

17/

Sam McDougle

• Agent57: Outperforming the human Atari benchmark

• Working Memory Graphs

• Hopfield Networks is All You Need

• Working memory is not a natural kind and cannot explain central cognition

• Toward an algorithmic theory of biology

18/

Kimberly Stachenfeld

• What Is a Cognitive Map? Organizing Knowledge for Flexible Behavior

• The hippocampus as a predictive map

• Failure to demonstrate replication and extension of Tolman et al.’s spatial learning experiment with humans

• Cognitive maps and spatial inference in animals

• Do animals have cognitive maps?

• Successor Uncertainties: Exploration and Uncertainty in Temporal Difference Learning

• Relational inductive biases, deep learning, and graph networks

• Compositionality

• The Tolman-Eichenbaum Machine

• Revisiting the global workspace orchestrating the hierarchical organization of the human brain

• The nested neural hierarchy and the self

• Complementary learning systems within the hippocampus

• The successor representation in human reinforcement learning

• Predictive representations can link modelbased reinforcement learning to model-free mechanisms

• Sharp wave ripples during learning stabilize hippocampal spatial map

• Adam Trischler

• Next steps toward human-level performance in neural language models

• Fine-Tuning GPT-2 from
  Human Preferences

• AI and Neuroscience: A virtuous circle

• Neuroscience-Inspired Artificial Intelligence

• Representing States & Spaces, CCN2019 Tutorial by Kim & Tim

• CCN 2019: Tutorial T-A: Representing states and spaces

• Moravec's paradox

• On the Measure of Intelligence

• Learning with an unreliable teacher

• Aligning Superhuman AI with Human Behavior: Chess as a Model System

• ROBUST CIRCUIT RHYTHMS IN SMALL CIRCUITS ARISE FROM VARIABLE CIRCUIT COMPONENTS AND MECHANISMS

• Dr Eve Marder - Differential Resilience to Perturbations of Circuits with Similar Performance

19/

Caswell Barry

• The place-cell representation of volumetric space in rats

• Grid cell firing fields in a volumetric space

• Organizing conceptual knowledge in humans with a gridlike code

• Learning Structures: Predictive Representations, Replay, and Generalization

• During hippocampal inactivation, grid cells maintain synchrony, even when the grid pattern is lost

• Grid cells require excitatory drive from the hippocampus

• Functional properties of stellate cells in medial entorhinal cortex layer II

• Lewis Carroll on mapping over the territory

• Impaired Spatial Representation in CA1 after Lesion of Direct Input from Entorhinal Cortex

• Functional connectivity of the entorhinal–hippocampal space circuit

• Transient optogenetic inactivation of the medial entorhinal cortex biases the active population of hippocampal neurons

• What do grid cells contribute to place cell firing?

• Patients with hippocampal amnesia cannot imagine new experiences

• Hippocampal Place Cells Encode Local Surface-Texture Boundaries

• Precise spatial representations in the hippocampus of a food‑caching bird

• Representation of edges, head direction, and swimming kinematics in the brain of freely-navigating fish

• Hippocampal Place Cells Encode Local Surface-Texture Boundaries

• Development of the Spatial Representation System in the Rat

• The self-organization of grid cells in 3D

• The 3D geometry of grid cells in flying bats

• The hippocampus as a predictive map

• Reinforcement Learning: An Introduction

• An Outsider's Tour of Reinforcement Learning

20/

Hugo Spiers

• Cities have a negative impact on navigation ability: evidence from 38 countries

• Predictive Maps in Rats and Humans for Spatial Navigation

• Hippocampal and prefrontal processing of network topology to simulate the future

• Hippocampal place cells construct reward related sequences through unexplored space

• Navigation-related structural change in the hippocampi of taxi drivers

• Human Spatial Navigation

• Hippocampal CA1 activity correlated with the distance to the goal and navigation performance

• Geometric models reveal behavioral and neural signatures of transforming naturalistic experiences into episodic memories

• Cerebellum Shapes Hippocampal Spatial Code

• Geometric models reveal behavioural and neural signatures of transforming experiences into memories

21/

Katja Hoffmann

• The Malmo Platform for Artificial Intelligence Experimentation

• AMRL: Aggregated Memory For Reinforcement Learning

• Challenges and Opportunities for Creating Agents in Commercial Games

• Designer-centered reinforcement learning

• Project Paidia

• Deep Interactive Bayesian Reinforcement Learning via Meta-Learning

• Project Malmo

• Synergistic Benefits of Group Search in Rats

• Understanding LSTM Networks

• Unsupervised Predictive Memory in a Goal-Directed Agent

• Optimizing Agent Behavior over Long Time Scales by Transporting Value

• For Stroke Patients, New Video Game May Help Restore More Function, Faster

• Dolphin App for Neurological Brain Function

• Pierre-Yves Oudeyer

• Exploring phylogeny to find the function of sleep

• The Temporal Dynamics of Opportunity Costs: A Normative Account of Cognitive Fatigue and Boredom

22/

Randall O'Reilly

• Deep Predictive Learning in Neocortex and Pulvinar

• emergent reboot in Go

• Simulating a Primary Visual Cortex at the Front of CNNs Improves Robustness to Image Perturbations

• Symbolic Behaviour in Artificial Intelligence

• Cerebellum Shapes Hippocampal Spatial Code

• Voxelwise encoding models show that cerebellar language representations are highly conceptual

• Understanding the Contribution of the Human Cerebellum across Task Domains

• Functional boundaries in the human cerebellum revealed by a multi-domain task battery

• The cerebellar cognitive affective/Schmahmann syndrome scale

• A new case of complete primary cerebellar agenesis: clinical and imaging findings in a living patient

• Motor Learning

• Pallidal neuron activity during sequential arm movements

• Next steps toward human-level performance in neural language models

• Learning to Combine Top-Down and Bottom-Up Signals in Recurrent Neural Networks with Attention over Modules

• Recurrent Independent Mechanisms

• Computational Cognitive Neuroscience Lectures by Randall O'Reilly

23/

Kirsty Graham

• How Nature Solves Problems Through Computation

• Our Inner Ape

• Meaning and ostension in great ape gestural communication

• Exorcising Grice's ghost: an empirical approach to studying intentional communication in animals

• The meanings of chimpanzee gestures

• Scratching beneath the surface: intentionality in great ape signal production

• The gestural repertoire of the wild bonobo (Pan paniscus): a mutually understood communication system

• A gestural repertoire of 1- to 2-year-old human children: in search of the ape gestures

• Just kidding: the evolutionary roots of playful teasing

• Interactional Foundations of Language: The Interaction Engine Hypothesis

• Rise of the Warrior Apes

• Bonobo and chimpanzee gestures overlap extensively in meaning

• Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning

• Paul Cisek: Backward Through The Brain

• The Origin of Concepts

• Theory of Mind: A Special Issue of Social Neuroscience

• Symbolic Behaviour in Artificial Intelligence

• Folk Physics for Apes: The Chimpanzee’s Theory of How the World Works

• A Turing-Like Handshake Test for Motor Intelligence

• Sapiens: A Brief History of Humankind 

• The manipulative complexity of Lower Paleolithic stone toolmaking

• Fractionating Human Intelligence

• Language and thought are not the same thing: evidence from neuroimaging and neurological patients

24/

Christopher Honey

• LEARNING REPRESENTATIONS FROM TEMPORALLY SMOOTH DATA

• The time course of perceptual choice: the leaky, competing accumulator model

• Constructing and Forgetting Temporal Context in the Human Cerebral Cortex

• Timescales of spontaneous fMRI fluctuations relate to structural connectivity in the brain

• Predictive Representations in Hippocampal and Prefrontal Hierarchies

• Predicting the Future with Multi-scale Successor Representations

• A nested cortical hierarchy of neural states underlies event segmentation in the human brain

• Representational content determines the long-term dynamics of the CA1 spatial code

• Individual Human Brain Areas Can Be Identified from Their Characteristic Spectral Activation Fingerprints

• Neuronal timescales are functionally dynamic and shaped by cortical microarchitecture

• The Prefrontal Cortex

• TRACX2: a connectionist autoencoder using graded chunks to model infant visual statistical learning

• The Similarity Structure of Real-World Memories

• Memory and Reality

• Using Smartphone, Social Media, and Sensor Data for Psychological Research

25/

Jacqueline Gottlieb

• Anxious individuals have difficulty learning the causal statistics of aversive environments

• Random Network Distillation: a new take on Curiosity-Driven Learning

• Give AI curiosity, and it will watch TV forever

• The Dark Room Problem

• Dorsal anterior cingulate-brainstem ensemble as a reinforcement meta-learner

• An Elephant in the Dark

• The expected value of control: An integrative theory of anterior cingulate cortex function

• Rats strategically manage learning during perceptual decision making

• An integrative theory of locus coeruleus-norepinephrine function: adaptive gain and optimal performance

• Increased locus coeruleus tonic activity causes disengagement from a patch-foraging task

• Neural correlates of dueling affective reactions to win–win choices

• Functional Connectivity under Anticipation of Shock: Correlates of Trait Anxious Affect versus Induced Anxiety

26/

Simon Kornblith

• A partition-based similarity for classification distributions

• An Information-Geometric Distance on the Space of Tasks

• Do ImageNet Classifiers Generalize to ImageNet?

• A Simple Framework for Contrastive Learning of Visual Representations

• SIMCLR - GitHub

• Invariant Information Clustering for Unsupervised Image Classification and Segmentation

• Rethinking supervised learning: insights from biological learning and from calling it by its name

• Exploring Simple Siamese Representation Learning

• Unsupervised neural network models of the ventral visual stream

• The Bitter Lesson

• Explaining and Harnessing Adversarial Examples

• Adversarial Patch

• Self-supervised learning through the eyes of a child

• Humans can decipher adversarial images

• What do adversarial images tell us about human vision?

• Performance vs. competence in human–machine comparisons

• Barlow Twins: Self-Supervised Learning via Redundancy Reduction

• Adversarial Examples that Fool both Computer Vision and Time-Limited Humans

• Large-scale, high-resolution comparison of the core visual object recognition behavior of humans, monkeys, and state-of-the-art deep artificial neural networks

• Direct Fit to Nature: An Evolutionary Perspective on Biological and Artificial Neural Networks

• Immediate susceptibility to visual illusions after sight onset

• Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet

• Hierarchical nucleation in deep neural networks