I am a fourth year graduate student working under the supervision of Karen Livescu at Toyota Technological Institute at Chicago (TTIC). I also work closely with Kevin Gimpel. Before TTIC, I spent two years as a software engineer in IBM Research, New Delhi. And even before that I spent the four most fun years of my life at Indian Institute of Technology Kanpur as an undergrad in Computer Science from 2009-13.
I am interested in machine learning and in particular, speech and natural language processing (NLP).
VibRein: An Engaging and Assistive Mobile Learning Companion for Students with Intellectual Disabilities
Shubham Toshniwal, Prasenjit Dey, Nitendra Rajput, Saurabh Srivastava
Australian Conference on Human-Computer Interaction (OzCHI) 2015
Visual information processing allocation between a mobile device and a network
Anirban Majumder, Samik Datta, Sharad Jaiswal, Nisheeth Shrivastava, Sreedal Menon, Shubham Toshniwal
U.S. Patent No. 8,913,838 B2, issued December 16, 2014
Recently, I have seen a trend of complex, deeper and sometimes unintuitive models claiming state-of-the-art results on some small/weird/self-created/new datasets. While I understand that empirical evidence has been the "ultimate test" in our field, I am not a big fan of such papers. It feels to me that this is the classic case of overfitting but over the sample of datasets. We need to be at guard against such ideas and possibly introduce some quality control measures.
I have a weird sense of humor and here are some catchphrases that I use and which make me tickle: