GDumb: A simple approach that questions our progress in continual learning
We discuss a general formulation for the Continual Learning (CL) problem for classification—a learning task where a stream provides samples to a learner and the goal of the learner, depending on the samples it receives, is to continually upgrade its knowledge about the old classes and learn new ones...
Main Authors: | Prabhu, A, Torr, PHS, Dokania, PK |
---|---|
Format: | Conference item |
Language: | English |
Published: |
Springer International Publishing
2020
|
Similar Items
-
RanDumb: a simple approach that questions the efficacy of continual representation learning
by: Prabhu, A, et al.
Published: (2025) -
RanDumb: random representations outperform online continually learned representations
by: Prabhu, A, et al.
Published: (2025) -
Continual learning in low-rank orthogonal subspaces
by: Chaudhry, A, et al.
Published: (2020) -
An embarrassingly simple approach to zero-shot learning
by: Romera-Paredes, B, et al.
Published: (2015) -
Progressive skeletonization: trimming more fat from a network at initialization
by: de Jorge, P, et al.
Published: (2020)