Ida Momennejad – Multi-scale Predictive Representations
Add to Calendar
When:
December 15, 2020 @ 12:00 pm – 1:00 pm
2020-12-15T12:00:00-05:00
2020-12-15T13:00:00-05:00
Abstract: Memory and planning rely on learning the relational structure of experience. A century after ‘latent learning’ experiments summarized by Tolman, the larger puzzle of cognitive maps remains elusive: how does the brain learn compact representations of relational structures to guide flexible behavior? I use reinforcement learning (RL) to study how humans learn predictive representations in memory and planning. I show behavioral, fMRI, and electrophysiology evidence that hippocampal and prefrontal hierarchies learn multi-scale predictive representations updated via offline replay. This approach advances the century old notion of cognitive maps and can inform biologically inspired artificial agents as well as computational psychiatry.
Bio: Ida Momennejad is a senior researcher in reinforcement learning at Microsoft Research NYC. Ida builds and tests neurally plausible algorithms for learning the structure of the environment such that it serves memory, exploration, & planning. Her approach combines reinforcement learning with behavioral experiments, fMRI, & electrophysiology. Ida got her BSc in software engineering (Tehran, Iran), MSc in Philosophy of Science (Utrecht, Netherlands), PhD in psychology/computational neuroscience (Berlin, Germany). She was a postdoc at Princeton and an associate research scientist at Columbia BME.