Thomas L Lee — accepted AI-conference papers

2025

Lightweight Online Adaption for Time Series Foundation Model Forecasts

Thomas L Lee, William Toner, Rajkarn Singh, Artjom Joosen, Martin Asenov

ICML 2025poster

Foundation models (FMs) have emerged as a promising approach for time series forecasting. While effective, FMs typically remain fixed during deployment due to the high computational costs of learning them online. Consequently, deployed FMs fail to adapt their forecasts to current data characteristic…

Cited by 1 Source ↗PDF ↗Scholar ↗

2024

Approximate Bayesian Class-Conditional Models under Continuous Representation Shift

Thomas L. Lee, Amos Storkey

AISTATS 2024poster

For models consisting of a classifier in some representation space, learning online from a non-stationary data stream often necessitates changes in the representation. So, the question arises of what is the best way to adapt the classifier to shifts in representation. Current methods only slowly cha…

Cited by 4 Source ↗PDF ↗code ↗Scholar ↗