Scholar
Gautham Vasan
Google Scholar ID: 4qMs0zMAAAAJ
University of Alberta, Amii
Citations & Impact
Citations
433
H-index
9
i10-index
9
Publications
19
Co-authors
18
list available
Publications
Resume (English only)
Academic Achievements
- NeurIPS 2024: 'Deep Policy Gradient Methods Without Batch Updates, Target Networks, or Replay Buffers' (first author)
- NeurIPS FitML Workshop 2024 (preprint): 'Streaming Deep Reinforcement Learning Finally Works'
- RLC 2024: 'Revisiting Sparse Rewards for Goal-Reaching Reinforcement Learning' (first author)
- CoRL MRM-D Workshop 2024: 'Versatile and Generalizable Manipulation via Goal-Conditioned Reinforcement Learning with Grounded Object Detection'
- AAMAS Doctoral Consortium 2024: 'Autonomous Skill Acquisition for Robots Using Graduated Learning' (first author)
- AAMAS 2024: 'MaDi: Learning to Mask Distractions for Generalization in Visual Deep Reinforcement Learning'
- ICML 2023: 'Correcting Discount-Factor Mismatch in On-Policy Policy Gradient Methods'
- Recipient of the Outstanding Thesis Award in Computing Science at the University of Alberta
Research Experience
- Deployed deep reinforcement learning to a fleet of warehouse robots at Kindred AI used by Gap Inc., developing the RLScan system for apparel barcode scanning
- Interned at Sanctuary AI working on imitation-bootstrapped reinforcement learning
- Interned at the University of Freiburg on imitation-guided RL projects
- Contributed to computational frameworks for real-time robot learning, including SenseAct and ReLoD
- Developed streaming RL algorithms: Action Value Gradient (AVG) and Stream-X
