Google Research

EvaNet: A Family of Diverse, Fast and Accurate Video Architectures

Bay Area Machine Learning Symposium (BayLearn) (2019)

Abstract

We present a novel evolutionary algorithm that automatically constructs architectures of layers exploring space-time interactions for videos. The discovered architectures are accurate, diverse and efficient. Ensembling such models leads to further accuracy gains and yields faster and more accurate solutions than previous state-of-the-art models. Evolved models can be used across datasets and to build more powerful models for video understanding.

Research Areas

Learn more about how we do research

We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work