Robust OT-Guided Generative Residual Domain Adaptation for Bike-Sharing Demand Prediction under Temporal Domain Shift
Yiming Ma · 2026
Read Robust OT-Guided Generative Residual Domain Adaptation for Bike-Sharing Demand Prediction under Temporal Domain Shift by Yiming Ma (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-23115.txt.
Description
Ma studies bike-sharing demand prediction under temporal domain shift, using March Citi Bike data from 2021 to 2026. Models trained on past station-hour data degrade as travel patterns change. The Gen-ROTDA framework fits a target station-time anchor from a labeled target subset. It transfers residuals rather than raw demand, generates residual features with a label-preserving generator, and trims costly matches before training the predictor. It achieves the lowest error on the 2025 to 2026 job and is the best optimal-transport-family method across years.
Catalog record
| CREATOR | Yiming Ma |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 21261 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.23115