Matthew T. Jackson

DPhil student at the University of Oxford

matthew_jackson.jpeg

I’m a third year DPhil student at the University of Oxford, where I am co-supervised by Jakob Foerster and Shimon Whiteson. I’m also a member of the AIMS CDT.

I’ve completed software engineering internships at Amazon, Arm and Cubica Technology, working on natural language processing, computer vision and efficient network inference. I previously completed an MSc in Machine Learning from University College London and a BA in Computer Science from the University of Cambridge.

Research

I’m interested in making agents learn faster than humans. Two promising approaches to this are learning from synthetic data and learning generalizable policies and algorithms with meta-reinforcement learning.

Policy-Guided Diffusion

Matthew T. Jackson, Michael T. Matthews, Cong Lu, Jakob Foerster, Shimon Whiteson

Work in progress - NeurIPS 2023 Workshop on Robot Learning

Discovering Temporally-Aware Reinforcement Learning Algorithms

Matthew T. Jackson*, Chris Lu*, Louis Kirsch, Robert T. Lange, Shimon Whiteson, Jakob N. Foerster

International Conference on Learning Representations (ICLR) 2024   paper

Discovering General Reinforcement Learning Algorithms with Adversarial Environment Design

Matthew T. Jackson, Minqi Jiang, Jack Parker-Holder, Risto Vuorio, Chris Lu, Gregory Farquhar, Shimon Whiteson, Jakob N. Foerster

Neural Information Processing Systems (NeurIPS) 2023   paper

Hypernetworks in Meta-Reinforcement Learning

Jake Beck, Matthew T. Jackson, Risto Vuorio, Shimon Whiteson

Conference on Robot Learning (CoRL) 2022   paper

Multi-Modal Fusion by Meta-Initialization

Matthew T. Jackson*, Shreshth A. Malik*, Michael T. Matthews, Yousuf Mohamed-Ahmed

arXiv preprint   paper