Sharan Vaswani

Sharan Vaswani is an Assistant Professor in the School of Computing Science at Simon Fraser University (SFU), Canada. His research focuses on designing algorithms for sequential decision-making under uncertainty, stochastic optimization, and their application to modern machine learning, particularly reinforcement learning and generalization analysis.

Education & Career:

• PhD in Computer Science (2015-2018), University of British Columbia (UBC), supervised by Laks Lakshmanan and Mark Schmidt.
• Postdoc (2019-2021) at Mila - Quebec AI Institute with Simon Lacoste-Julien, and University of Alberta with Csaba Szepesvári.
• MSc in Computer Science (2015), UBC, focusing on influence maximization in social networks.
• BS from Birla Institute of Technology and Science, Pilani (2012), followed by research engineering at Siemens Corporate Research.

Research Interests:

• Algorithmic development for decision-making in uncertain environments (bandits, reinforcement learning).
• Stochastic optimization methods with provable guarantees.
• Generalization and theoretical foundations of machine learning models.

Recent Trends in Publications:

• Focus on optimization algorithms (e.g., stochastic gradient methods, line search, momentum techniques) with theoretical analysis.
• Contributions to reinforcement learning, including policy gradient methods and constrained MDPs.
• Exploration of adaptive algorithms for continual learning and over-parameterized models.

Awards:

• Best Paper Honorable Mention (AISTATS 2022).
• Best Paper Award (2nd IEEE International Conference on Parallel Distributed and Grid Computing 2012).

Research Group: Leads a team at SFU focused on machine learning optimization and decision-making systems. Active in organizing workshops at NeurIPS and ICML on optimization and reinforcement learning theory.