Events

Transformers learn in-context by (functional) gradient descent Xiang Cheng, TILOS Postdoctoral Scholar at MIT HDSI 123 and Zoom: https://ucsd.zoom.us/j/99334315002 Abstract: Motivated by the in-context learning phenomenon, we investigate how the Transformer neural network can implement learning algorithms in its forward pass. We show that a linear Transformer naturally learns to implement gradient descent, which enables it to […]

The Triton Neurotech and TNT Academy team is excited to announce their first event in their Professor Talk series! Join them for a talk put on by Dr. Bradley Voytek on Wednesday, May 8th from 7-8pm in Henry Booker Room, Jacobs Hall 2512. Dr. Voytek is a Professor in the Department of Cognitive Science, the Halıcıoğlu […]

Abstract: Recent progress in AI can be attributed to the emergence of large models trained on large datasets. However, teaching AI agents to reliably interact with our physical world has proven challenging, which is in part due to a lack of large and sufficiently diverse robot datasets. In this talk, I will cover ongoing efforts of the Open X-Embodiment project–a collaboration between 279 researchers across 20+ institutions–to build a large, open dataset for real-world robotics, and discuss how this new paradigm is rapidly changing the field. Concretely, I will discuss why we need large datasets in robotics, what such datasets may look like, and how large models can be trained and evaluated effectively in a cross-embodiment cross-environment setting. Finally, I will conclude the talk by sharing my perspective on the limitations of current embodied AI agents, as well as how to move forward as a community.

Speaker: Sarah H. Cen Abstract: We have begun grappling with difficult questions related to the rise of AI, including: What rights do individuals have in the age of AI? When should we regulate AI and when should we abstain? What degree of transparency is needed to monitor AI systems? These questions are all concerned with AI accountability: determining who owes responsibility and to whom in […]

Abstract: Although influenza is one of the best-studied viruses, vaccine effectiveness remains around 20-50%. A sizable fraction of people exhibit a weak or short-lived antibody response following vaccination, yet we cannot identify these individuals a priori nor ascertain whether a different vaccine would have served them better. In this talk, we demonstrate how machine learning can […]