Sherjil Ozair

Hi! I graduated with a bachelors and masters degree in Computer Science and Engineering, from Indian Institute of Technology Delhi in 2015. I am interested in artificial intelligence applications to solve real-world problems.

Currently

I'm currently working at Baidu Research's Silicon Valley AI lab at Sunnyvale, California. Here, I work on deep learning for speech recognition. In particular, I worked on neural language modeling for speech decoding, and low-latency small-footprint trigger word detection. My current research interest is semi-supervised domain (accent, noise, etc.) adaption for speech recognition.

Previously

Prof. Mausam and I did a semi-instructional Neural Networks and Deep Learning Study Group at IIT Delhi.

I've been a visiting researcher at Prof. Yoshua Bengio's LISA Lab, where I worked on Unsupervised Deep Learning, in particular Generative Stochastic Networks and Generative Adversarial Networks among others.

I interned at Microsoft Research, Bangalore in Summer 2013 with Aditya Nori and Sriram Rajamani where I worked on Probabilistic Programming.

In my first year summer, I was selected for Google Summer of Code 2011. I worked on Sympy, and initiated its sparse linear algebra package.

Drop me a line!

Publications

Efficient Synthesis of Probabilistic Programs, Aditya Nori, Sherjil Ozair, Sriram Rajamani, Deepak Vijaykeerthy, PLDI 2015

Deep Directed Generative Autoencoders, Sherjil Ozair, Yoshua Bengio, NIPS 2014 Deep Learning Workshop

Generative Adversarial Networks, Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio, NIPS 2014

On the Equivalence Between Deep NADE and Generative Stochastic Networks, Li Yao, Sherjil Ozair, Kyunghyun Cho, Yoshua Bengio, ECML 2014

Multimodal Transitions for Generative Stochastic Networks, Sherjil Ozair, Li Yao, Yoshua Bengio, ICLR 2014 Workshop