method · first seen 22 Oct 2026
knowledge distillation regularisation
Constraining a model to match earlier model outputs to limit forgetting when learning new tasks.
Mentions per week
PapersCodeDiscussion
Earliest sources
- REDAR: Reinforcement-Driven Adaptive Regularisation for Class Incremental LearningPapers · 22 Oct 2026 · Akila Ayanthi, Darshana Priyasad, Tharindu Fernando, Sridha Sridharan, Clinton F…
Latest sources
- REDAR: Reinforcement-Driven Adaptive Regularisation for Class Incremental LearningPapers · 22 Oct 2026 · Akila Ayanthi, Darshana Priyasad, Tharindu Fernando, Sridha Sridharan, Clinton F…