← back to home

CV / Portfolio

Temporally-organized overview of some my work.

This page will be updated soon…

Projects

• min-llm: (WIP) Minimal implementations of LLM components for learning purposes. 2024- 

• GPbeT: An LLM-powered user interface to assist in sports betting. 2024 

• CS 231N Project: Prototypical pre-training for visual representation learning. 2023 

• CS 330 Project: Gradient-based meta learning for morphologically diverse few-shot cell segmentation. 2022 

• CS 221 Project: Distributionally robust multi-task optimization for fair skin cancer classification. 2022 

• torchplate: A minimal, open-source experiment framework for machine learning workflows in PyTorch. 2022-2023 

• rsbox: An open-source toolbox of utility functions for machine learning in Python. 2022-2023 

• ddpm-pytorch: Minimal implementation of denoising diffusion probabilistic models (DDPM) in PyTorch. 2022-2023 

• CS 109 Project: Developed a novel approach using Gaussian distributions to model curvature in hand-drawn spirals. Won CS 109 challenge (1st place among 239 participants) 2022 

• lemnos: An open source light-weight command-line to-do list manager, built with python. 2021-2022 

Research

• “Contrastive ground-level image and remote sensing pre-training improves representation learning for natural world imagery” (ECCV 2024).
• “Beyond the imitation game: Quantifying and extrapolating the capabilities of language models” (ICLR 2023, Spotlight).
• “Prototypical Pre-Training for Visual Representation Learning”