Siddarth Venkatraman

I am a PhD student at Mila, Quebec AI Institute affiliated with Université de Montréal where I am exploring ways to scale Reinforcement Learning with large offline datasets for robotics. I'm particularly interested in the idea of iterative optimization as a form of reasoning, and learnt policies which are capable of solving complex problems efficiently are best viewed as amortized optimizers. I am advised Dr. Glen Berseth and part of the Robotics and Embodied AI Lab (REAL).

I obtained my Masters in Robotics from Carnegie Mellon University where I was a part of the Auton Lab and the Argo AI Center for Autonomous Vehicles Research advised by Dr. Jeff Schneider. Before this I obtained by bachelors in Computer Science at Manipal Institute of Technology. I worked as an ML research intern at NASA Jet Propulsion Laboratory (JPL) where I worked on learning heuristics for more efficient Mars Rover motion planning.

I am generally interested in topics related to reinforcement learning, representation learning and generative models. I am looking for collaborators on research projects related to these topics. If you are interested in working together, please reach out by email!

Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn  

profile photo

Research

Here is a collection of my previous research work.

rtb Amortizing intractable inference in diffusion models for vision, language and control

Siddarth Venkatraman*, Moksh Jain*, Luca Scimeca*, Minsu Kim*, Marcin Sendera*, Mohsin Hasan, Luke Rowe, Sarthak Mittal, Pablo Lemos, Emmanuel Bengio, Alexandre Adam, Jarrid Rector-Brooks, Yoshua Bengio, Glen Berseth, Nikolay Malkin
arXiv preprint

ldcq Reasoning with Latent Diffusion in Offline Reinforcement Learning

Siddarth Venkatraman*, Shivesh Khaitan*, Ravi Tej Akella*, John Dolan, Jeff Schneider, Glen Berseth
International Conference on Learning Representations (ICLR 2024)

oposm Learning Temporally Abstract World Models without Online Experimentation

Benjamin Freed, Siddarth Venkatraman, Guillaume Adrien Sartoretti, Jeff Schneider, Howie Choset
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:10338-10356, 2023. (ICML 2023)

alphasac Multi-Alpha Soft Actor-Critic: Overcoming Stochastic Biases in Maximum Entropy Reinforcement Learning

Conor Igoe, Swapnil Pande, Siddarth Venkatraman, Jeff Schneider
2023 IEEE International Conference on Robotics and Automation (ICRA 2023)

mlnav MLNav: Learning to Safely Navigate on Martian Terrains

Shreyansh Daftry , Neil Abcouwer, Tyler Del Sesto, Siddarth Venkatraman, Jialin Song, Lucas Igel, Amos Byon, Ugo Rosolia, Yisong Yue, Masahiro Ono
IEEE Robotics and Automation Letters ( Volume: 7, Issue: 2, April 2022) (RAL+ICRA 2022)

mlnav2 Machine Learning Based Path Planning for Improved Rover Navigation

Neil Abcouwer, Shreyansh Daftry, Tyler Del Sesto, Olivier Toupet, Masahiro Ono, Siddarth Venkatraman, Ravi Lanka, Jialin Song, Yisong Yue
2021 IEEE Aerospace Conference (50100)

steg Deep Residual Neural Networks for Image in Audio Steganography (Workshop Paper)

Shivam Agarwal*, Siddarth Venkatraman*
2020 IEEE Sixth International Conference on Multimedia Big Data (BigMM)