Physics of Learning and Adaptation I
FOCUS · MAR-J58 · ID: 3112548
Presentations
-
High variance weight updates and sampling dynamics strongly affect how diffusion generative models generalize: a path-integral view
ORAL
–
Publication: Generalization through variance: how noise shapes inductive biases in diffusion models (ICLR 2025 submission)
Presenters
-
John Joseph Vastola
- Harvard Medical School
Authors
-
John Joseph Vastola
- Harvard Medical School
-
-
Parameter Symmetry and Noise Equilibrium of Stochastic Gradient Descent
ORAL
–
Presenters
-
Liu Ziyin
- Massachusetts Institute of Technology
Authors
-
Liu Ziyin
- Massachusetts Institute of Technology
-
Mingze Wang
- Peking University
-
Hongchao Li
- University of Tokyo
-
Lei Wu
- Peking University
-
-
Understanding and Optimizing Gradients of Neural Networks Using Dynamical Systems Theory
ORAL
–
Presenters
-
Rainer Engelken
- Columbia University
Authors
-
Rainer Engelken
- Columbia University
-
Larry F Abbott
- Columbia University
-
-
Tracing a naturalistic artificial neural network's adaptation through memory and prediction
ORAL
–
Presenters
-
Abigail M Minin
- Scripps College
Authors
-
Sarah Marzen
- Scripps, Pitzer & Claremont McKenna College
-
Nicol Harper
- University of Oxford
-
Abigail M Minin
- Scripps College
-
Kate Rabinowitz
- Claremont McKenna College
-
Devon Xiong
- Pitzer College
-
-
Resource-rational reinforcement learning and sensorimotor causal states, \\ and resource-rational maximiners
ORAL
–
Presenters
-
Sarah Marzen
- Scripps, Pitzer & Claremont McKenna College
Authors
-
Sarah Marzen
- Scripps, Pitzer & Claremont McKenna College
-
-
Recurrent neural networks balance sensory- and memory-guided policies for spatial foraging.
ORAL
–
Presenters
-
Scott Sterrett
- University of Washington
Authors
-
Scott Sterrett
- University of Washington
-
David H Gire
- University of Washington
-
Adrienne Fairhall
- University of Washington
-
-
'Interacting' parametric models of learning: insights into the adaptation to data
ORAL · Invited
–
Presenters
-
Yasaman Bahri
- Google DeepMind
Authors
-
Yasaman Bahri
- Google DeepMind
-
-
Emergent discreteness in optimal behavior
ORAL
–
Presenters
-
Michael C Abbott
- Yale University
Authors
-
Michael C Abbott
- Yale University
-
Benjamin B Machta
- Yale University
-
-
Phase transitions induced by resource limitations in biological information processing
ORAL
–
Publication: Takehiro Tottori and Tetsuya J. Kobayashi, "Resource Limitations induce Phase Transitions in Biological Information Processing", arXiv:2409.14002 (2024)
Takehiro Tottori and Tetsuya J. Kobayashi, "Theory for Optimal Estimation and Control under Resource Limitations and Its Applications to Biological Information Processing and Decision-Making", arXiv:2409.14003 (2024)Presenters
-
Takehiro Tottori
- RIKEN Center for Brain Science
Authors
-
Takehiro Tottori
- RIKEN Center for Brain Science
-
Tetsuya J Kobayashi
- The University of Tokyo
-
-
Fixational eye movements and retinal adaptation: optimizing motion to maximize information acquisition.
ORAL
–
Publication: Fixational Eye Movements and Detection Thresholds, Alexander J.H. Houston, David H. Brainard, Hannah E. Smithson, Daniel J. Read (in preparation)
Presenters
-
Daniel J Read
- University of Leeds
Authors
-
Daniel J Read
- University of Leeds
-
Alexander J Houston
- University of Glasgow
-
David H Brainard
- University of Pennsylvania
-
Fabian Coupette
- University of Leeds
-
Hannah E Smithson
- University of Oxford
-
-
Analyzing an organism's sensors using Maximum Entropy models with bias, variance, and confusion matrices
ORAL
–
Presenters
-
Christopher Wang
- Pomona College
Authors
-
Sarah Marzen
- Scripps, Pitzer & Claremont McKenna College
-
Christopher Wang
- Pomona College
-
Elianna Schimke
- Scripps College
-
Martina Lamberti
- University of Twente
-
Joost le Feber
- University of Twente
-
Tristan Kako
- Pioneer Research Program
-