Tom Rossmann

Applied Data Scientist

Applied mathematician building production ML systems. Focus on rigor, transparency, and mathematical foundations.

NLP · Time Series · Anomaly Detection

Projects

YahtzeeRL

Yahtzee agent built with JAX/Flax/RLax, trained through self-play with stochastic MuZero-style MCTS and explicit chance nodes for dice rolls.

81.1% win rate vs heuristic baseline over 512 games

Solar Irradiance Forecasting

Solar irradiance forecasting as a multi-horizon task, comparing GRU, LSTM, seasonal naive, and persistence baselines.

GRU reduced RMSE by ~34% at 4h and ~21% at 24h vs strongest naive baseline

Thesis

Kernel Methods for Regression

Kernel methods for regression problems by implicitly mapping input variables into high-dimensional feature spaces. Comparison of standard ridge regression and kernel ridge regression with algebraic equivalence proof.