Daniel Poh

DPhil Student, Engineering, Machine Learning Research Group

Daniel's research explores Gaussian Processes and Reinforcement Learning. With the latter topic, he is particularly interested in data efficiency, knowledge transfer and continuous learning for agents in the context of market microstructure.

Daniel has recently written a paper with Martin Tegner and Stephen Roberts demonstrating the efficacy of Gaussian Processes for commodity risk management. He is currently assisting Jan-Peter Calliess with work relating to decision making for agents.

 

Related Research: Data Analysis and Patterns in Data, Electronic Trading, Decision Making under Uncertainty and Asset Allocation