Research & Projects
Papers
A Basin-Selection Perspective on Grokking via Singular Learning Theory
This paper analyses grokking through the lens of Singular Learning Theory. In particular, it frames grokking as a phase transition between near-zero-loss basins with different geometry.
GenGNAR: Genetic Graph Search for GNAR Models with Applications to Inflation Forecasting
The paper looks at genetic algorithms for graph selection in Generalised Network Autoregressive models; and applies the resulting algorithm to UK inflation forecasting.
Miscellaneous work
PyTorch Experimentation
A repository of notebooks where I explore PyTorch-based models, from linear baselines to more advanced RAG-oriented models.
Part III Essay: Topological Data Analysis
A Part III essay on topological data analysis, covering core ideas and methods from computational topology and their use in data analysis.
Economics Bachelor Thesis: Modern Monetary Theory in Japan
This thesis was part of the requirements for UNED's Bachelor Thesis in Economics. It studies whether Japan serves as a good example of a state which has applied the doctrines of Modern Monetary Theory (MMT). Throughout it, we look at the basic precepts of MMT, the macroeconomic situation of Japan in the last decades and a brief history of its monetary and fiscal policies to answer the thesis' question.
Other Projects
RL x LLM Reading Club (Imperial College London)
I am setting up a bi-weekly reading club with Paul Lezeau at Imperial to explore reinforcement learning and large language models, from core concepts (RLHF, transformer architecture and variants, LLM egineering, etc) to current topics such as reasoning and planning.
The format is discussion-oriented and shaped by participant interests. Researchers and students with a basic ML background are welcome. Email me if you are interested in joining!