Home | Sitemap | Contact | Chinese | CAS
Search: 
About AMSS Research People International Cooperation News Societies & Journals Papers Resources Education & Training Join Us Links
Research
Location: Home >  Research >  Colloquia & Seminars
(2017.11.10 10:00 N420)Scott Reed:Recent Advances in Autoregressive Density Estimation with Deep Neural Networks
Author:
ArticleSource:
Update time: 2017-11-17
Close
A A A
Print
 

 

Academy of Mathematics and Systems Science, CAS
Colloquia & Seminars

Speaker:

Scott Reed, Research Scientist, Deepmind

Inviter:  
Title:
Recent Advances in Autoregressive Density Estimation with Deep Neural Networks
Time & Venue:
2017.11.10 10:00-11:30 N420
Abstract:

Autoregressive models parametrized as deep neural networks (called PixelCNN) achieve state-of-the-art results in image density estimation. Although training is fast, sampling is costly, requiring one network evaluation per pixel; O(N) for N pixels. In this talk, I will describe a parallelized PixelCNN that achieves competitive density estimation and orders of magnitude speedup - O(log N) sampling instead of O(N) - enabling the practical generation of high-resolution images. Results will be presented on class-conditional image generation, text-to-image synthesis, and action-conditional video generation. In the second part of the talk, I will discuss density estimation in the low-data regime as a meta learning problem.

Scott is a research scientist at DeepMind. He completed his PhD with Honglak Lee at the University of Michigan in 2016. His research focuses on deep learning methods for image generation, object detection, imitation learning and program induction.

 

 

Appendix:
Copyright@2008, All Rights Reserved, Academy of Mathematics and Systems Science, CAS
Tel: 86-10-62553063 Fax: 86-10-62541829 E-mail: contact@amss.ac.cn