My name is Oli Bridge. I'm a Master's student at UCL Computer Science studying machine learning. I previously graduated with a BA/MSci from the University of Cambridge, where I researched path integral quantum dynamics in my final year at the Yusuf Hamied Department of Chemistry.
- 🌱 My research interests broadly include reinforcement learning and probabilistic machine learning
- 🔭 I am currently working in the Machine Intelligence Lab at UCL, researching symmetries in graph reinforcement learning
- 📫 Reach me at: olibridge@rocketmail.com
- 👨🏻🎓 My Google Scholar
TeXOCR is an Optical Character Recognition (OCR) model for converting images of equations into LaTeX code. This project contains a fully built transformer-based OCR model, along with a FastAPI web application to showcases its abilities.
DeepRL is an implementation of various model-free deep reinforcement learning algorithms built with PyTorch. The project utilises OpenAI's gymnasium environments to train the algorithms, as well as produced rendered outputs to visualise the agent's performance.
MaskedImageModelling is a PyTorch implementation of a self-supervised learning framework to pre-train a model for an image segmentation task. A Vision Transformer is pre-trained by learning to reconstruct partially masked images, and fine-tuned on the Oxford-IIIT Pets dataset for segmentation.