Pith. sign in

REVIEW 5 cited by

Towards Robust Evaluations of Continual Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1805.09733 v3 pith:3GMQBVNW submitted 2018-05-24 stat.ML cs.LG

classification stat.MLcs.LG
keywords learningcontinualevaluationsexperimentresearchapproachesdesideratadesigns
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Experiments used in current continual learning research do not faithfully assess fundamental challenges of learning continually. Instead of assessing performance on challenging and representative experiment designs, recent research has focused on increased dataset difficulty, while still using flawed experiment set-ups. We examine standard evaluations and show why these evaluations make some continual learning approaches look better than they are. We introduce desiderata for continual learning evaluations and explain why their absence creates misleading comparisons. Based on our desiderata we then propose new experiment designs which we demonstrate with various continual learning approaches and datasets. Our analysis calls for a reprioritization of research effort by the community.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Online Continual Learning with Maximally Interfered Retrieval

    cs.LG 2019-08 accept novelty 7.0 of 10

    Selecting replay samples by estimated loss increase after a virtual update improves online continual learning performance over random replay.

  2. Fine-Tuning Regimes Define Distinct Continual Learning Problems

    cs.LG 2026-04 unverdicted novelty 6.0 of 10

    The relative rankings of continual learning methods are not preserved across different fine-tuning regimes defined by trainable parameter depth.

  3. Lifelong Machine Learning with Deep Streaming Linear Discriminant Analysis

    cs.LG 2019-09 conditional novelty 6.0 of 10

    A fixed-feature CNN plus streaming linear discriminant analysis achieves strong continual classification performance with orders of magnitude less memory and compute than rehearsal-based incremental learners.

  4. Toward Understanding Catastrophic Forgetting in Continual Learning

    cs.LG 2019-08 conditional novelty 6.0 of 10

    Error rates of continual learning algorithms on a task sequence correlate strongly with the sum of individual task complexities, but not with the dissimilarity between consecutive tasks.

  5. Autoencoder-Based Hybrid Replay for Class-Incremental Learning

    cs.LG 2025-05 reject novelty 5.0 of 10

    AHR combines an autoencoder with a repulsive-force class centroid layout to replay compressed latent exemplars, reporting higher accuracy than selected CIL baselines.

Pith tools