AI Paper Atlas

Research map / Continual learning / Online and replay-based

Learning curves for continual learning in neural networks: Self-knowledge transfer and forgetting

Ryo Karakida, Shotaro Akaho

ICLR 2022 · Continual learning › Online and replay-based · approach: Theoretical analysis of forgetting · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Train sequentially on task A before task B, with target similarity above the critical valueNTK continual learning · Subsequent task Bgeneralization error: E_{A\rightarrow B}/E_B \sim 2(1-\rho); \rho>1/2Improved
Train sequentially on task A before task B when target similarity is sufficiently lowNTK continual learning · Subsequent task Bgeneralization error: \rho < \rho^*:= \sqrt{\gamma_A}/(1+\sqrt{\gamma_A})Hurt
Train on task B after task A when targets are even subtly dissimilarNTK continual learning · Previous task Abackward generalization error: E_{A\rightarrow B}^{back}(\rho) \sim \sum_i \eta_i^2 (1-\rho)Hurt
Use equal sample sizes for the two tasks with the same targetNTK continual learning · Same target, two tasksgeneralization error: E_{A\rightarrow B}(1) < E_{ave} < E_A=E_BImproved
Make task B's sample size much larger than task A's for the same targetNTK continual learning · Same target, task Bgeneralization error relative to task-B-only training: E_{A\rightarrow B}(1)/E_B \sim 1/(1-\gamma_A)Hurt
Make task B's sample size much smaller than task A's for the same targetNTK continual learning · Same target, previous task Ageneralization error relative to task-A-only training: E_{A\rightarrow B}(1)/E_A \sim 1/(1-\gamma_B)Hurt
Train on additional tasks with equal sample sizes and the same targetNTK continual learning · Same target, multiple tasksgeneralization error: E_{n+1} \leq E_n; if all eigenvalues are positive, E_{n+1} < E_nImproved
Add label noiseSequential training on the same target · Synthetic data and deep-network experimentsgeneralization errorHurt
Use unbalanced sample sizes across tasksSequential training on the same target · Synthetic data, MNIST, and CIFAR-10generalization errorMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy