This CSH talk by Rahim Entezari will take place online from 15:00—16:00 on May 29 2020.
If you are interested in participating, please email office@csh.ac.at in order to receive information on how to access the online talk.
Abstract:
While deep networks are a highly successful model class they require substantial computation resources for their training, storage, and inference, which limits their effective use on resource-constrained devices. Many recent research activities explore different options for compressing and optimizing deep models for the Internet of Things (IoT). Our work so far has focused on two important aspects of deep neural network compression: class-dependent model compression and explainable compression. In this talk, we shortly discuss why these aspects are important for real-world applications and summarize our contributions.