Hello! I'm Anup. I have joined David R. Cheriton School of Computer Science at the University of Waterloo from Fall 2019 where I am working under the supervision of Prof. Ming Li as a graduate student. I am also fortunate to work with Prof. Lili Mou affiliated with AMII and Department of Computing Science at University of Alberta. I completed my bachelors degree from IIIT-B where I was affiliated to the Multimodal Perception lab, led by Prof. Dinesh Babu

I enjoy working on deep learning methods that are especially applied to Recommender Systems and Natural Language Processing. You can find details of my Publications and Projects here.

I am currently working on unsupervised neural methods and reinforcement learning for parsing human languages. Without using the human-labeled constituency trees or other syntactic labeling, we can force our model to learn the dependanices in a natural way and possibly mimic the way humans learn the language.

Within Recommender Systems (RS) my research focuses on building efficient and scalable Machine Learning algorithms which go beyond accuracy. During my thesis at IIIT-B, broadly, my goal was to exploit generative models to learn user-item interaction function in RS.

  • [August 2020] Simsketch - A simple and efficient sketching algorithm for dimensionality reduction is accepted in ACML 2020!
  • [March 2020] Work on semantic similarity and text retrieval is up on arXiv
  • [May 2019] Successfully defended my undergrad thesis at IIIT-B
  • [Feb 2019] Gave my first TEDx pre-event talk titled "Burst your Filter Bubble"
  • [Sept 2018] Two papers accepted in ACML and ICDM! My first small though significant contribution to ML research :)

Apart from my work, I enjoy travelling and hiking. I am also an avid reader of world politics and get considerable enjoyment from sports.