לוח שנה משולב

How does one adapt a pre-trained visual model to novel downstream tasks without task-specific fine-tuning or changing the model weights? Inspired by prompting in NLP, Visual Prompting is an exciting new paradigm that guides computer vision models to perform downstream tasks by providing visual examples. This talk will focus on designing the underlying self-supervised learning objectives, architectures, and various methods to prompt vision models. Finally, I will discuss in-context learning from a mechanistic interpretability standpoint and share new insights into how transformer models can encode task-specific information in their activation space.

Bio: Amir Bar is a PhD student at Tel Aviv University and Berkeley AI Research. During his PhD, Amir has focused on self-supervised learning in computer vision and pioneered Visual Prompting (or visual in-context learning). Before starting his PhD, he led the research team at the healthcare startup Zebra Medical Vision (recently acquired), where his team developed multiple FDA-approved algorithms now in clinical use in hospitals worldwide. Starting in August 2024, Amir will join FAIR Labs as a postdoctoral researcher, working with Yann LeCun.

Color each vertex of an infinite graph blue with probability p and red with probability 1-p, independently among vertices. For which values of p is there an infinite connected component of blue vertices? The talk will focus on this classical percolation problem for the class of planar graphs. Recently, Itai Benjamini made several conjectures in this context, relating the percolation problem to the behavior of simple random walk on the graph. We will explain how partial answers to Benjamini's conjectures may be obtained using the theory of circle packings. Among the results is the fact that the critical percolation probability admits a universal lower bound for the class of recurrent plane triangulations. No previous knowledge on percolation or circle packings will be assumed.

The mixing time and spectral gap of a random walk on the symmetric group can sometimes be understood in terms of its low dimensional representations (e.g., Aldous' spectral gap conjecture). It turns out that under a mild degree condition involving the step of the group, the same holds for nilpotent groups w.r.t. their one dimensional representations: the spectral gap and the epsilon total variation mixing time of the walk on G are determined by those of the projection of the walk to the abelianization G/[G,G]. We'll discuss some applications concerning the cutoff phenomenon (= abrupt convergence to equilibrium) and the dependence (or lack of!) of the spectral gap and the mixing time on the choice of generators.  

As time permits we shall discuss a related result, confirming in the nilpotent setup a conjecture of Aldous and Diaconis concerning the occurrence of cutoff when a diverging number of generators are picked uniformly at random. Joint work with Zoe Huang.