Alexander Imani Cowen-Rivers
Research Engineer
MediaGamma
I was a postgraduate student studying Machine Learning & Data Science at University College London, advised by Prof. Sebastian Riedel and Dr. Pasquale Minervini in the Machine Reading Group. My main area of interest is learning with structured data and structured representations/computations. These topics include natural language processing, multi-agent reinforcement learning and logic.
I completed my BSc at the University of Bath, studying Mathematical Sciences. My first two years I studied mathematics, and in my final year of studies, I changed course to studying mathematics with computer science. After my penultimate year of study, I completed a 12 month placement at Burberry PLC , as a Data Analyst.
News
- 12/20: We Won "NeurIPS 2020 Black-box Optimisation Challenge" with : "HEBO: Heteroscedastic Evolutionary Bayesian Optimisation". Code coming soon.
- 12/20: We released: "Are we Forgetting about Compositional Optimisers in Bayesian Optimisation?" on arxiv. Code coming soon.
- 10/20: Best Systems Paper + Long Oral at CoRL: "SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving" accepted at CoRL 2020. Code
- 2/20: We released the paper "Compositional ADAM: An Adaptive Compositional Solver" on arxiv.
- 1/20: We released the paper "Safe Model-Based & Active Reinforcement Learning" on arxiv.
- 6/19: I began a research engineer role with Huawei Noahs Ark London.
- 12/19: Spotlight "Emergent Communication with World Models" at NeurIPS 2019, Emergent Communication Worksop.
- 6/19: I began an internship with Preferred Networks.
- 4/19: Together with For.ai, we have released a highly modularised RL codebase. Paper and Code.
- 12/18: I began collaborating with For.ai.
- 10/18: I joined MediaGamma as a Research Engineer, reporting partly into Prof. Jun Wang.
- 5/18: Our work on Towards Incremental Cylindrical Algebraic Decomposition in Maple was accepted (Oral) at FLoC 2018.
- 5/18: I received a student travel award to present our work at the Third International Workshop on Satisfiability Checking and Symbolic Computation, University of Oxford, FLoC 2018.
- 12/17: I published a technical report Lazard Projection and Lift Scheme, which details the classical and incremental algorithms for Lazard's algorithm
- In the summer of 2017, I worked as a Research Assistant as Coventry University, I also attended the Second International Workshop on Satisfiability Checking and Symbolic Computation.
