Model Based Inference of Synaptic Plasticity Rules
Primary Area: applications to neuroscience & cognitive science
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Keywords: synaptic plasticity, neuroscience, biologically plausible learning, model fitting
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
TL;DR: Introduce an efficient framework to learn synaptic plasticity rules from recorded neural activity and/or behavior. We apply our framework to behavioral data from fruit flies, revealing an active forgetting mechanism during plasticity.
Abstract: Understanding learning through synaptic plasticity rules in the brain is a grand challenge for neuroscience. Here we introduce a novel computational framework for inferring plasticity rules from experimental data on neural activity trajectories and behavioral learning dynamics. Our methodology parameterizes the plasticity function to provide theoretical interpretability and facilitate gradient-based optimization. For instance, we use Taylor series expansions or multilayer perceptrons to approximate plasticity rules, and we adjust their parameters via gradient descent over entire trajectories to closely match observed neural activity and behavioral data. Notably, our approach can learn intricate rules that induce long time-dependencies, such as those incorporating weight decay. We validate our method through simulations, accurately recovering established rules, like Oja's, and more complex hypothetical rules incorporating reward-modulated terms. We assess the resilience of our technique to noise and, as a tangible application, apply it to behavioral data from \emph{Drosophila} during a probabilistic reward-learning experiment. Remarkably, we identify an active forgetting component of reward learning in flies that enhances the predictive accuracy of previous models. Overall, our modeling framework provides an exciting new avenue to elucidate the computational principles governing synaptic plasticity and learning in the brain.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 3073
Loading