SAMARTH SINHA
samarth [dot] sinha [at] mail [dot] utoronto [dot] ca
I am currently at MILA as a visiting student with Anirudh Goyal, Hugo Larochelle and Yoshua Bengio.
I am a fourth year undergraduate student in Computer Engineering at the University of Toronto where I work with Animesh Garg and Roger Grosse. Previously, I was a visiting student at UC Berkeley, working with Trevor Darrell and Zeynep Akata; before I worked with Ozan Sener as a part of CVGL at Stanford University. I have also been very fortunate to work at some amazing startups: Aurora, DeepMap, Concured, Nanoleaf.
I spent my high school summers working with Alex Lvovsky working on Quantum Optics and with Guenther Ruhe working on Software Engineering.
RESEARCH INTERESTS
My research interests span across Deep Learning and Reinforcement Learning. I am interested in training models that can learn with minimal supervision, while generalizing beyond the test set.
I have had the privilege to work with some amazing people, and I am always interested in chatting about cool problems. If you find anything interesting on the website, please get in touch!
PUBLICATIONS / PRE-PRINTS
CONSISTENCY REGULARIZATION FOR VARIATIONAL AUTOENCODERS
Samarth Sinha, Augustus Odena, Adji Dieng (2020)
In review, arXiv soon
DIVA: DIVERSE VISUAL FEATURE AGGREGATION FOR DEEP METRIC LEARNING
Karsten Roth*, Timo Milbich*, Homanga Bharadhwaj, Samarth Sinha, Joseph Paul Cohen, Bjorn Ommer, Yoshua Bengio (2020)
In review, arXiv soon
CURRICULUM BY TEXTURE
Samarth Sinha, Animesh Garg, Hugo Larochelle (2020)
TOP-K TRAINING OF GANS: IMPROVING GAN PERFORMANCE BY THROWING AWAY BAD SAMPLES
Samarth Sinha, Anirudh Goyal, Colin Raffel, Augustus Odena (2020)
EXPERIENCE REPLAY WITH LIKELIHOOD-FREE IMPORTANCE WEIGHTS
Samarth Sinha*, Jiaming Song*, Animesh Garg, Stefano Ermon (2020)
In review, arXiv soon
REVISITING TRAINING STRATEGIES AND GENERALIZATION PERFORMANCE IN DEEP METRIC LEARNING
Karsten Roth*, Timo Milbich*, Samarth Sinha, Prateek Gupta, Bjoern Ommer, Joseph Paul Cohen (2020)
DIBS: DIVERSITY INDUCING INFORMATION BOTTLENECK IN MODEL ENSEMBLES
Samarth Sinha*, Homanga Bharadhwaj*, Anirudh Goyal, Hugo Larochelle, Animesh Garg, Florian Shkurti (2020)
Samarth Sinha, Han Zhang, Anirudh Goyal, Yoshua Bengio, Hugo Larochelle, Augustus Odena (2019)
Samarth Sinha*, Sayna Ebrahimi*, Trevor Darrell
ICCV 2019 (Oral)
Edgar Schönfeld, Sayna Ebrahimi, Samarth Sinha, Trevor Darrell, Zeynep Akata
CVPR 2019, ICLR LLD Workshop 2019