Matthew Morris

Matthew Morris
Wolfson Building, Parks Road, Oxford OX1 3QD
Interests
My research interests are broadly within the field of neuro-symbolic artificial intelligence, with a particular focus on neuro-symbolic logic models. However, I also have a history of working with both pure logic and reinforcement learning.
Biography
I did my undergraduate degree at the University of Cape Town in Mathematics and Computer Science, followed by a Master’s degree at Oxford in Computer Science. I have received numerous academic awards, was a Harry Crossley fellow and a Skye Foundation Scholar, and am currently funded by the EPSRC. I am currently doing my DPhil (PhD) in Computer Science at Oxford, supervised by Ian Horrocks, and am focusing on neuro-symbolic logic models. My general area of interest is anything neuro-symbolic. In the course of my career, I have worked as a software developer for Amazon Web Services, a research assistant, and most recently as a research engineer for InstaDeep, where I worked on multi-agent reinforcement learning research. I have several publications in the areas of machine learning, logic, and reinforcement learning, most recently in NeurIPS where I used Graph Neural Networks for communication in multi-agent reinforcement learning. If you want to play some ultimate frisbee, challenge me to a chess match, or chat about my research, please send me an email.
Selected Publications
-
Relational Graph Convolutional Networks Do Not Learn Sound Rules
Matthew Morris‚ David J Tena Cucala‚ Bernardo Cuenca Grau and Ian Horrocks
In 21st International Conference on Principles of Knowledge Representation and Reasoning. 2024.
Details about Relational Graph Convolutional Networks Do Not Learn Sound Rules | BibTeX data for Relational Graph Convolutional Networks Do Not Learn Sound Rules
-
Orbit−Equivariant Graph Neural Networks
Matthew Morris‚ Bernardo Cuenca Grau and Ian Horrocks
In The Twelfth International Conference on Learning Representations. 2024.
Details about Orbit−Equivariant Graph Neural Networks | BibTeX data for Orbit−Equivariant Graph Neural Networks | Link to Orbit−Equivariant Graph Neural Networks
-
Universally Expressive Communication in Multi−Agent Reinforcement Learning
Matthew Morris Thomas D Barrett and Arnu Pretorius
In Thirty−sixth Conference on Neural Information Processing Systems (NeurIPS). 2022.
Details about Universally Expressive Communication in Multi−Agent Reinforcement Learning | BibTeX data for Universally Expressive Communication in Multi−Agent Reinforcement Learning | Link to Universally Expressive Communication in Multi−Agent Reinforcement Learning