Algorithmic Learning

Algorithmic learning focuses on creating systems that learn from data. Analysis of these systems are done through the lens of their predictive performance, as well as their time and space complexity. While this view point includes diverse approaches to learning like PAC learning, reinforcement learning, and kernel learning, my primary focus is on online learning, where algorithms adapt and learn sequentially from incoming data.

Game-Theoretic Learning

Game-theoretic learning is about environments of strategic interaction between multiple decision-makers, or "players." My research has concentrated on zero-sum repeated games, analysing the dynamic interplay between inputs, outputs, and the learning function within these competitive environments.

Algorithmic Game-theoretic Learning

My core interest lies at the intersection of Algorithmic & Game-theoretic view points. I design and analyse algorithms that can be framed as games, prioritising solutions that exhibit provably bounded regret, low space complexity, and high computational efficiency. This synthesis allows for the creation of robust and adaptive systems capable of strategic decision-making.

Applications

My research focuses on the real-world applications of adaptive and intelligent systems. I explore these concepts further in two ways: On my blog, I analyse US equities and ETFs, while in the Projects section, I document the practical implementation of my ongoing work. For blog access, please send a request via email.