method · first seen 26 Oct 2026
flow Q-learning
Also: FQL
Offline and online RL method that uses expressive flow-based policies trained with value guidance.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Freeze, Don't Clone: Fine-Tuning with Expressive Policies Without Retaining Offline DataPapers · 26 Oct 2026 · Sami Nur Islam, Johan Obando-Ceron, Doina Precup, Aaron Courville, Pablo Samuel…
Latest sources
- Freeze, Don't Clone: Fine-Tuning with Expressive Policies Without Retaining Offline DataPapers · 26 Oct 2026 · Sami Nur Islam, Johan Obando-Ceron, Doina Precup, Aaron Courville, Pablo Samuel…