From Neuroscience to Artificially Intelligent Systems
April 5 - 9, 2022

You must register for the meeting in order to submit abstracts.
After registering you will be sent a web link for abstract submission by email.
You may copy and paste your abstract from Word, Google Docs, or Notepad; abstracts are limited to ~2900 characters.

Program information: An electronic version of the program abstract book will be sent three working days prior to the first day of the meeting, and hard copies will be available for collection upon your arrival at Cold Spring Harbor Laboratory. First night and keynote speakers are informed of their session date and time, otherwise program information is only available upon release of the electronic version of the abstract book. The reason we do this is to try and maximize interactions by encouraging participants to stay for the duration of the meeting.

Please check your email for talk length, poster instructions, and how to have your poster printed at CSHL for collection upon arrival. 

Abstract Status

Presenting Author

Abstract Title

Talk/Poster

Adeli Jelodar, Hossein

A brain-inspired object-based attention network for multi-object recognition and visual reasoning

talk

Ahn, Seoyoung

Reconstruction-as-feedback serves as an effective attention mechanism to increase recognition robustness

talk

Aldarondo, Diego E

A virtual rodent predicts the structure of neural activity across natural behavior

talk

An, Su Jin

Learning state-space uncertainty, but not value uncertainty, is sufficient for metacognitive exploration

poster

Aqrabawi, Afif

Knowledge is abstracted across overlapping memory engrams

poster

Aquino, Tomas

Neurons in human pre-supplementary motor area encode key computations for value-based choice

poster

Barabasi, Daniel L

Emergence of functional circuits in the absence of neural activity

talk

Barto, Andrew

Can machine learning tell us anything about the neural basis of learning?

talk

Bennett, James

Learning and forgetting from the perspective of control

poster

Bhattasali, Nikhil

Neural circuit architectural priors for embodied control

talk

Biderman, Dan

Lightning Pose—A suite of semi-supervised networks for robust video tracking with minimal manual annotation

poster

Bouttier, Vincent

Better inferences by decorrelating information with circular belief propagation

poster

Bricken, Trenton

Sparse distributed memory, top-K networks, and relations to deep learning

poster

Burger, Thomas

Active dendrites support reliable spiking computations

poster

Can, Tankut

Emergence of memory manifolds

talk

Chandra, Sarthak

Architectural biases for computer vision from developmental processes for the visual cortex

poster

Chavlis, Spyridon

Empowering artificial neural networks by adding biological dendrites

poster

Chen, Chao B

A topological profiling of neural networks and its application to detection of backdoor attacks

poster

Chen, Guozhang

Multiplexing of robust visual processing capabilities by a data-based cortical microcircuit model

poster

Chen, Yusi

Hippocampus as a generative circuit for predictive coding of future sequences

poster

Conwell, Colin

Computational evidence for implicit embedding of visually evoked affect in feedforward, purely perceptual machines

poster

Coon, William G

Continual learning with three sleep components—NREM, REM, and synaptic downscaling

poster

Czegel, Daniel

Darwinian neurodynamics

poster

De Silva, Laknath A

Kernel density networks

poster

Dey, Jayanta

Out-of-distribution detection using kernel density polytopes

poster

Dowling, Matthew

Dimension-free embedding for meta-learning latent dynamics

poster

Eckmann, Samuel

Unsupervised competitive Hebbian learning explains the emergence of functional recurrent E-I networks

poster

Ehrlich, David A

How difficult is it to access information in neural networks?

poster

Fehrman, Christof

The use of biologically realistic spectro-temporal receptive fields produces noise-invariant spiking neural networks

poster

Ferreira Castro, André

Mechanistic model of effective pattern separation in a memory and learning circuit

poster

Gardner, Daniel

What do neural circuits compute? How? What should AI emulate?

talk

Garibbo, Michele

What deep reinforcement learning tells us about human motor learning and vice-versa

poster

Genkin, Alexander

Biological learning of identity preserving transformations

poster

George, Dileep

Space is a sequence—A unifying model for spatial, temporal, and relational abstraction in hippocampal cognitive maps

talk

Gershman, Sam

Amortized inference in mind and brain

talk

Golkar, Siavash

A neuron as an Eigen-solver of time-series dynamics

poster

Greedy, William H

Single-phase and efficient deep learning in cortico-cortical networks

poster

Gurbuz, Mustafa B

NISPA—Neuro-inspired stability-plasticity adaptation for continual learning in sparse networks

poster

Haider, Paul

Latent equilibrium—A unified learning theory for arbitrarily fast computation with arbitrarily slow neurons

poster

Han, Yena

Training as a confound—Insights into biological and artificial model comparisons with untrained networks

poster

Holzhausen, Konstantin

Robustness of neural networks trained with biologically inspired local learning rules

poster

How, Javier J

Larval zebrafish demonstrate transfer learning

poster

Humphries, Joseph

Selective attention using a temporal difference method for uncertainty learning in a valence-partitioned reinforcement learning framework

poster

Ito, Takuya

Multi-task representations in human cortex transform along a sensory-to-motor hierarchy

poster

Ivanitskiy, Michael I

Robustness in reinforcement learning through unconstrained dreaming

poster

Janik, Romuald A

The visual brain seen through the lens of a deep neural network

poster

Kepple, Daniel R

Curriculum learning as a tool to uncover learning principles in the brain

poster

Kerg, Giancarlo

On neural architecture inductive biases for relational tasks

talk

Kim, Christopher

Transfer of learned neural dynamics through unstructured connectivity in strongly coupled spiking networks

poster

Kohan, Adam

Forward learning for spiking neural networks

talk

Konkle, Talia

Emergence of object topography motifs from self-organization of a unified object space

poster

Kornblith, Simon

Understanding artificial neural networks through their hidden representations

talk

Kosoy, Eliza

Learning causal overhypotheses through exploration in children and computational models

poster

Koulakov, Alexei

Genomic bottleneck approach to faster learning

talk

Kowadlo, Gideon

Why do brains have 2 hemispheres and would it improve deep learning?—A study with left and right neural networks

poster

Kudithiipudi, Dhireesha

Exploring complex synapses for task-agnostic continual learning accelerators

poster

Lafond-Mercier, Raphael

Networks of adaptive neurons used in spatial learning

poster

Lepperod, Mikkel E

Manifold propagation through recurrent neural networks

poster

Liang, Tong

Evolution of innate behavioral strategies through competitive population dynamics

poster

Libedinsky, Camilo

A transient subspace for eye-movement initiation in the primate prefrontal cortex

poster

Lindsey, Jack W

Computational principles of systems memory consolidation

poster

Lindsey, Jack W

Incorporating prior knowledge in reinforcement learning—A basal ganglia-inspired model

poster

Lipshutz, David

Reverse engineering early visual processing using fly connectomics data

poster

Liu, Yuhan Helena

Beyond accuracy—Robustness and generalization properties of biologically plausible learning rules

talk

Logiaco, Laureline

Brain insights improve RNNs’ accuracy and robustness for hierarchical control of continually learned autonomous motor motifs

poster

Luna, Raúl

An intrinsically nonlinear motion model predicts different visual phenomena that cannot be explained by classical energy models without incorporating additional stages

poster

Maass, Wolfgang

How does structure induce function in cortical microcircuits?

talk

Makino, Hiroshi

Arithmetic value representation for hierarchical behavior composition

poster

Mansinghka, Vikash

Brain computation as fast spiking neural Monte Carlo inference in probabilistic programs

poster

Marschall, Owen E

Meta-dynamics of recurrent neural networks during learning—General patterns and algorithmic signatures

poster

Masset, Paul

A distributional code for learning across timescales in dopamine-based reinforcement learning

poster

Meulemans, Alexander

The least-control principle for learning

poster

Mollard, Sami

Recurrent neural networks for learning multi-step visual routines with reinforcement learning

talk

Moore, Jasmine

Simulating neurodegeneration with noise in convolutional neural networks

poster

Morales, Guillermo B

"Smooth" input representations in ANNs stem from critical dynamics.

poster

Nagy, David G

Episodic memory and generative replay for online structure learning

poster

Nejatbakhsh, Amin

Controlled switching linear dynamics—A framework for causal interrogation of neural networks

poster

Nguyen, Tri M

Structured connectivity in the cerebellum enables noise-resilient neural network

poster

Nowak, Aleksandra I

Analyzing neural networks based on random graphs

poster

Orban, Gergo

Visual cortex-inspired hierarchical deep generative model architecture of natural images

talk

Panda, Priyadarshini

Rethinking the role of ANN-oriented techniques for spiking neural networks

talk

Patel, Devdhar

Robust control in real time reinforcement learning using adaptive response times

poster

Pedamonti, Dabal

Hippocampal networks support continual learning and generalisation

poster

Pedigo, Benjamin D

Towards statistical comparative connectomics—A case study on the bilateral symmetry of an insect brain connectome

poster

Pemberton, Joseph O

A cerebellar solution to cortical temporal credit assignment

poster

Pemberton, Joseph O

Cortico-cerebellar networks as decoupling neural interfaces

poster

Pettersen, Markus

Studying emergent place cells and remapping in a recurrent neural network model

poster

Pettine, Warren W

Human latent-state generalization through prototype learning with discriminative attention

poster

Piera Lindez, Pau

Adversarial and variational autoencoders improve metagenomics binning

poster

Pir Hayatifard, Delaram

Understanding the olfactory code in awake mice

poster

Podolak, Igor T

Neuron ensembles in SNN networks

poster

Pouget, Alexandre

Talking nets—A neural model for generalization with natural language instructions

talk

Raghavan, Guruprasad

Engineering flexible machine learning systems inspired by biological intelligence

talk

Raj, Rishabh

RCNet—A biologically inspired network for unsupervised capturing of input redundancy and robust object representation

poster

Rajan, Kanaka

Curriculum learning to probe learning principles in biological and artificial networks

talk

Rawat, Shivang

Resonance in stochastic dynamical system models of cortical circuits

poster

Robinson, Brian S

Angular path integration with a ring attractor composed of synapse-constrained motifs

poster

Robinson, Brian S

Dimensionality reduction and point attractor dynamics in a recurrent connectome-constrained insect learning center model

poster

Roelfsema, Pieter

BrainProp—Solving the credit assignment problem in the brain

talk

Rogge, Frederik

On environmental sensitivity of the successor representation

poster

Rungratsameetaweemana, Nuttida

Flexible hierarchical computation in task-driven information processing

poster

Safron, Adam

Dream to explore—5-HT2a as adaptive temperature parameter for sophisticated affective inference

poster

Sainath, Pravish

Recurrent neural networks trained on N-back task reveal the functional activity of visual working memory

poster

Sands, Paul

Reward is not enough—A valence-partitioned reinforcement learning framework for dopamine-based control of human choice and affect dynamics

poster

Schaeffer, Rylan

Streaming inference for nonstationary clustering

poster

Schapiro, Anna

Learning representations of specifics and generalities over time

talk

Schneider, Andreas C

Infomorphic networks—Locally learning neural networks derived from partial information decomposition

poster

Schoyen, Vemund S

Place and grid cell navigation in multiple environments

poster

Schwartz, Alexander

Training recurrent activity in sparsely connected spiking neural networks

poster

Sedigh-Sarvestani, Madineh

What we lose by modeling the visual system without topographic maps

talk

Segert, Simon

Function learning and extrapolation via the principle of maximum entropy

poster

Sharma, Sugandha

Content addressable memory without catastrophic forgetting by heteroassociation with a fixed scaffold

talk

Shin, Jong M

Novel artificially intelligent machine manifests human-like behavior

poster

Shuvaev, Sergey

A normative theory of social hierarchy

poster

Tamari, Ronen

Evaluating situation modelling in reading comprehension in humans and machines

poster

Tesileanu, Tiberiu

A more realistic predictive coding model of cortical hierarchy

poster

Trattner, Jonathan

Parallel processing of temporal-difference prediction errors for emotionally evocative stimuli predicts human choice behavior and associated subjective feelings

poster

Tresp, Volker

A unified theory of perception, memory and semantic decoding

poster

Tsodyks, Misha

Human memory—Theory vs experiments

talk

Tuladhar, Anup

Changes in internal representational structure within a degenerating neural network

poster

Turcu, Denis

Analysis of electrosensing using an active electrolocation framework

poster

Turcu, Denis

Numerical cognition does not guarantee arithmetic ability

poster

van den Berg, Alexandra R

Biologically plausible gated recurrent neural networks for working memory and learning-to-learn

poster

Virgili, Samuele

Mind the gradient—Context-dependent selectivity to natural images in the retina revealed with a novel perturbative approach

poster

Vo, Vy A

Using neuroscience to improve long-range sequential processing in language models

poster

Westerberg, Jacob A

Attentional priority computed and precisely modified in sensory cortex

poster

Wu, Yujie

Brain-inspired global-local learning for versatile learning scenarios

poster

Xie, Marjorie

Task-dependent optimal representations in cerebellum-like networks

poster

Xie, Yudi

Models of flexible working memory with naturalistic stimuli

poster

Xu, Haoyin

Simplest streaming trees

poster

Yamashiro, Kotaro

Deep learning-based decoding of surface textures from rat somatosensory cortical activity

poster

Yin, Yijie

Stereotypy of network architecture of comprehensive synapse-resolution connectomes of //Drosophila melanogaster//

poster

Yiu, Eunice

Symmetry preference in 3D object completion

talk

Yu, Yiyi

A similarity neuron network with hubs—Underlying functional circuit organization of mouse visual cortex

poster

Zenke, Friedemann

Information processing with artificial spiking neural networks

talk

Zhang, Ruiyi

Inductive bias of neural architectures for spatial navigation

poster

Zylerberg, Joel

Photoreceptor biophysics enable deep learning models to generalize across light levels

poster