Algorithmic Learning

Algorithmic learning focuses on creating systems that learn from data, analyzed through the lens of predictive performance, time complexity, and space complexity. This perspective encompasses diverse approaches including PAC learning, reinforcement learning, and kernel methods. I am particularly interested in online learning, where algorithms adapt and learn sequentially from streaming data.

Game-Theoretic Learning

Game-theoretic learning studies environments of strategic interaction between multiple decision-makers, or "players." My research focuses on zero-sum repeated games, analyzing the dynamic interplay between inputs, outputs, and learning functions within these competitive settings. This framework provides powerful insights into adversarial learning scenario.

Algorithmic Game-Theoretic Learning

My work synthesizes algorithmic and game-theoretic perspectives to design and analyze learning algorithms that can be framed as strategic games. I prioritize solutions that exhibit provably desirable predictive, space, and time complexities while maintaining strategic robustness. This interdisciplinary approach enables the creation of adaptive systems capable of principled decision-making in competitive environments.

AI via Algorithmic Game-theoretic Learning

Broadly speaking, my work falls under the umbrella of AI. I design and analyze algorithms that learn.