Participants will learn about advanced techniques and models that allow for the disentanglement of visual factors in images and the compositionality of these factors to produce more meaningful representations. Disentanglement and Composition are believed to be one of the possible ways for AI to fundamentally understand the world, and eventually achieve Artificial General Intelligence (AGI). Specifically, we will cover the following topics:Session #1: Disentangled Representation Learning (DRL); Session #2: Latent Semantics Discovery; Session #3: Disentanglement and Equivariance; Session #4: Composition and Disentanglement for AGI.

Please contact Xin Jin for general inquiries.