Gautam Singh

singh.gautam@rutgers.edu
https://scholar.google.com/citations?user=lXpFxDwAAAAJ
https://twitter.com/gautamsi11 Github
  I am a Ph.D. student in Computer Science at Rutgers University focusing on Machine Learning, advised by Prof. Sungjin Ahn

The research questions I am interested in are:
  • How can we infer structured representations of data without supervision? 
  • Since organisms are endowed with innate priors, what kind of innateness can we bake into our models to achieve this?
  • Do the inferred representations also have some causal interpretation?
In exploring these, I rely on building probabilistic models using variational approximations and deep learning. 


Prior to starting a Ph.D. in 2018, I worked at IBM Research India for 3 years after obtaining a B.Tech. from IIT Guwahati in 2015
PUBLICATIONS   2020
Robustifying Sequential Neural Processes
Jaesik Yoon, Gautam Singh, Sungjin Ahn
ICML 2020

SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decomposition
{Zhixuan Lin, Yi-Fu Wu, Skand Peri, Weihao Sun}, Gautam Singh, Fei Deng, Jindong Jiang, Sungjin Ahn
ICLR 2020

2019
Sequential Neural Processes [pdf, code, project, slides, poster]
{Gautam Singh, Jaesik Yoon}, Youngsung Son, Sungjin Ahn
NeurIPS 2019  Spotlight