Ciara Pike-Burke

Ciara Pike-Burke is a Lecturer in Statistics at the Department of Mathematics, Imperial College London, within the Faculty of Natural Sciences. Her research focuses on statistical machine learning, particularly sequential decision making under uncertainty, including multi-armed bandits, reinforcement learning, and online learning. She holds a PhD from Lancaster University (STOR-i program) and was a postdoc at Universitat Pompeu Fabra in Barcelona.

Education:

• PhD in Statistics (Lancaster University, 2019)
• Postdoctoral Researcher (Universitat Pompeu Fabra, 2019-2022)

Research Interests: She explores theoretical aspects of sequential decision problems in areas like education and healthcare, emphasizing algorithmic efficiency and regret minimization. Her work addresses challenges such as delayed feedback, sparse rewards, and privacy constraints in bandit and reinforcement learning frameworks.

Publications: Her recent work includes advancements in hierarchical reinforcement learning, bandit algorithms with delayed feedback, and privacy-preserving techniques. Notable contributions include the QuACK algorithm for cooperative bandits and theoretical analyses of optimistic exploration strategies.

Outreach: She develops educational tools like the TryBandits web-based game to introduce students to statistical concepts. She actively recruits PhD students in reinforcement learning theory, bandits, and online learning.