**Summary:** While the fields of neuroscience and (especially) artificial intelligence have rapidly progressed over the last decade, what is perhaps less appreciated is how intertwined these two fields have become. This presentation will discuss two studies that highlight the synergy between these fields of research. In the first part, we will discuss how recurrent neural network models can be used to generate novel hypotheses regarding the neural substrates of working memory in the brain. These network models suggest that the maintenance and manipulation of information in working memory engage different neural substrates, potentially addressing a long-standing debate in the field. In the second part (paper attached), we will discuss how insights from neuroscience can be used to create novel algorithms that allow neural networks to learn large numbers of tasks with little forgetting. We show that mimicking the brain's strategy of employing multiple, complementary methods can be a highly efficient and effective approach to alleviate forgetting in neural networks.
Thursday, 1/17/2019
11:00am-12:20pm
Kent 120
A light lunch will be provided by Jimmy John's.
**Nicolas Masse** is a staff scientist in the Department of Neurobiology at the University of Chicago. His research focuses on developing neuroscience-inspired artificial intelligence, and is primarily interested in creating neural networks that can seamlessly switch between different contexts and environments. Before arriving in Chicago, Nicolas worked on developing brain-computer interface systems for people with tetraplegia at Brown University, researched the olfactory system at Cambridge, and studied the neural mechanisms underlying visual motion perception at McGill University. Prior to academia, Nicolas developed statistical models for a sports gambling company, and taught high school mathematics.
The 2018-2019 Computational Social Science Workshop meets Thursdays from 11 a.m. to 12:20 p.m. in Kent 120. All interested faculty and graduate students are welcome.
Students in the Masters of Computational Social Science program are expected to attend and join the discussion by posting a comment on the issues page of the workshop's public repository on GitHub. Further instructions are documented in the Computational Social Science Workshop's README on Github.