Timezone: »
Poster
The Neural Costs of Optimal Control
Samuel J Gershman · Robert C Wilson
Optimal control entails combining probabilities and utilities. However, for most practical problems probability densities can be represented only approximately. Choosing an approximation requires balancing the benefits of an accurate approximation against the costs of computing it. We propose a variational framework for achieving this balance and apply it to the problem of how a population code should optimally represent a distribution under resource constraints. The essence of our analysis is the conjecture that population codes are organized to maximize a lower bound on the log expected utility. This theory can account for a plethora of experimental data, including the reward-modulation of sensory receptive fields.
Author Information
Samuel J Gershman (Harvard University)
Robert C Wilson (Princeton University)
More from the Same Authors
-
2021 : CCNLab: A Benchmarking Framework for Computational Cognitive Neuroscience »
Nikhil Bhattasali · Momchil Tomov · Samuel J Gershman -
2023 Poster: Successor-Predecessor Intrinsic Exploration »
Changmin Yu · Neil Burgess · Samuel J Gershman · Maneesh Sahani -
2022 : Closing Remarks »
Samuel J Gershman -
2022 Workshop: Information-Theoretic Principles in Cognitive Systems »
Noga Zaslavsky · Mycal Tucker · Sarah Marzen · Irina Higgins · Stephanie Palmer · Samuel J Gershman -
2022 : Panel Discussion: Opportunities and Challenges »
Kenneth Norman · Janice Chen · Samuel J Gershman · Albert Gu · Sepp Hochreiter · Ida Momennejad · Hava Siegelmann · Sainbayar Sukhbaatar -
2018 Poster: Human-in-the-Loop Interpretability Prior »
Isaac Lage · Andrew Ross · Samuel J Gershman · Been Kim · Finale Doshi-Velez -
2018 Spotlight: Human-in-the-Loop Interpretability Prior »
Isaac Lage · Andrew Ross · Samuel J Gershman · Been Kim · Finale Doshi-Velez -
2016 Poster: Probing the Compositionality of Intuitive Functions »
Eric Schulz · Josh Tenenbaum · David Duvenaud · Maarten Speekenbrink · Samuel J Gershman -
2015 Workshop: Bounded Optimality and Rational Metareasoning »
Samuel J Gershman · Falk Lieder · Tom Griffiths · Noah Goodman -
2014 Poster: Design Principles of the Hippocampal Cognitive Map »
Kimberly Stachenfeld · Matthew Botvinick · Samuel J Gershman -
2014 Spotlight: Design Principles of the Hippocampal Cognitive Map »
Kimberly Stachenfeld · Matthew Botvinick · Samuel J Gershman -
2009 Poster: Perceptual Multistability as Markov Chain Monte Carlo Inference »
Samuel J Gershman · Edward Vul · Josh Tenenbaum -
2009 Spotlight: Perceptual Multistability as Markov Chain Monte Carlo Inference »
Samuel J Gershman · Edward Vul · Josh Tenenbaum -
2009 Poster: A Neural Implementation of the Kalman Filter »
Robert C Wilson · Leif Finkel -
2009 Poster: A Bayesian Analysis of Dynamics in Free Recall »
Richard Socher · Samuel J Gershman · Adler Perotte · Per Sederberg · David Blei · Kenneth Norman