Neural Nets for Cognition @ CogSci 2022

We are hosting the 3rd annual Neural Networks for Cognition discussion group this year at CogSci 2022! The goal of this event is to facilitate the exchange of ideas among researchers who are interested in using neural networks to understand human cognition. Our theme this year will be: "How can we tell how well a neural network is emulating human behavior?" We find this question especially timely, given recent high-profile releases from AI companies touting new successes of neural network-based AI systems and ongoing efforts to determine the relevance of these systems for understanding human cognition. We are organizing two events, to accommodate in-person as well as virtual participation. First, on July 19th at 1:00 PM - 2:00 PM EDT, we will be hosting a virtual panel discussion over Zoom, featuring the following invited guests: - Ellie Pavlick, Brown & Google - Ishita Dasgupta, Deepmind - Grace Lindsay, UCL & soon to be Asst. Prof @ NYU - Jacob Andreas, MIT We will be using Zoom to host the virtual discussion panel. Please register using the following link if you plan to attend: https://purdue-edu.zoom.us/meeting/register/tJUtd-urqD0oGNMulEOvFQfN5f9tCOC9WiGU Next, on July 28th at 4:40 PM - 6:15 PM EDT, we will be hosting an in-person discussion group at the Metro Toronto Convention Centre. This event will begin with a brief follow-up to the panel discussion, followed by our usual practice of breaking out into smaller groups organized by theme to continue the conversation. Please sign up here if you would like to join the online plenary session and/or the in-person discussion group during the conference. You will be able to edit your responses after submitting, to update your interests or other information later.