Skip to yearly menu bar Skip to main content


Oral

A Generalization Theory for Zero-Shot Prediction

Ronak Mehta · Zaid Harchaoui

West Ballroom A
[ ] [ Visit Oral 3B Representations 1 ]
Wed 16 Jul 10 a.m. — 10:15 a.m. PDT

Abstract:

A modern paradigm for generalization in machine learning and AI consists of pre-training a task-agnostic foundation model, generally obtained using self-supervised and multimodal contrastive learning. The resulting representations can be used for prediction on a downstream task for which no labeled data is available. We present a theoretical framework to better understand this approach, called zero-shot prediction. We identify the target quantities that zero-shot prediction aims to learn, or learns in passing, and the key conditional independence relationships that enable its generalization ability.

Chat is not available.