Dhaivat Bhatt
Research Masters student at MILA, Universite de Montreal
Office: 2259, Pavillon Andre Aisenstadt
2920 Ch de la Tour
Montreal, Quebec H3T 1N8
I am a Research masters student at Montreal Institute For Learning Algorithms (MILA), advised by Prof. Liam Paull. I finished my undergraduation from Birla institute of technology and science, Pilani in Electronics and instrumentation in 2016. I was a Masters by research student at Robotics research center, IIIT Hyderabad with Prof. K Madhava Krishna.
During my masters at IIIT Hyderabad, I extensively worked on problems like road boundary/curb detection and intersection localization using graphical models and Deep learning. For my thesis, I worked on solving stochastic control sequence problem in a Model Predictive control framework.
During my research at MILA, I worked on various projects such as Deep active localization and maplite. I joined in as a Masters student with Liam Paull's group and since then I have worked on wide range of problems including Out of distribution detection, probabilistic object detection and Generative classifiers. I have diverse set of interests, ranging from Deep learning, Computer vision, Robotics and Optimal control.
When I am not doing research, I am probably doing a course on history or debating politics with someone.
news
July 7, 2020 | Our position paper on probabilistic object detection has been accepted to AI for autonomous driving at ICML 2020. |
Jan 22, 2020 | Maplite has been accepted to RAL and will be presented to ICRA 2020. |
Sept 2019 | Joined Research masters program at MILA - Quebec AI institute under Professor Liam Paull. |
Jul 12, 2019 | Deep active localization has been accepted to RAL 2019. |
Jul 5, 2019 | Paper accepted to RO-MAN 2019. |
May, 2019 | Volunteered as a student at ICRA 2019, held at Montreal, Quebec. |
Nov 29, 2018 | Attending NeurIPS 2018. |
Nov 20, 2018 | Joined MILA - Quebec AI institute as a visiting researcher. |
Jun 15, 2017 | Paper accepted in IROS 2017! |
Jun, 2016 | Paper accepted in ICVGIP 2016! |