Pith. sign in

REVIEW 5 cited by

EMP-SSL: Towards Self-Supervised Learning in One Training Epoch

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 2304.03977 v1 pith:VY4GX7OU submitted 2023-04-08 cs.CV cs.AI

classification cs.CVcs.AI
keywords learningself-supervisedemp-sslmethodtrainingepochscifar-10cifar-100
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Recently, self-supervised learning (SSL) has achieved tremendous success in learning image representation. Despite the empirical success, most self-supervised learning methods are rather "inefficient" learners, typically taking hundreds of training epochs to fully converge. In this work, we show that the key towards efficient self-supervised learning is to increase the number of crops from each image instance. Leveraging one of the state-of-the-art SSL method, we introduce a simplistic form of self-supervised learning method called Extreme-Multi-Patch Self-Supervised-Learning (EMP-SSL) that does not rely on many heuristic techniques for SSL such as weight sharing between the branches, feature-wise normalization, output quantization, and stop gradient, etc, and reduces the training epochs by two orders of magnitude. We show that the proposed method is able to converge to 85.1% on CIFAR-10, 58.5% on CIFAR-100, 38.1% on Tiny ImageNet and 58.5% on ImageNet-100 in just one epoch. Furthermore, the proposed method achieves 91.5% on CIFAR-10, 70.1% on CIFAR-100, 51.5% on Tiny ImageNet and 78.9% on ImageNet-100 with linear probing in less than ten training epochs. In addition, we show that EMP-SSL shows significantly better transferability to out-of-domain datasets compared to baseline SSL methods. We will release the code in https://github.com/tsb0601/EMP-SSL.

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. A Principled Framework for Multi-View Contrastive Learning

    cs.LG 2025-07 conditional novelty 6.0 of 10

    Two multi-view contrastive losses, MV-InfoNCE and MV-DHEL, match InfoNCE's asymptotic optimum while improving accuracy and embedding rank with more views.

  2. Replay-free Online Continual Learning with Self-Supervised MultiPatches

    cs.LG 2025-02 conditional novelty 5.0 of 10

    Continual MultiPatches (CMP) trains self-supervised online continual learning models on multiple patches of each single example, reaching higher linear probing accuracy than most experience replay baselines without st...

  3. Seeing the Whole in the Parts in Self-Supervised Representation Learning

    cs.LG 2025-01 conditional novelty 5.0 of 10

    CO-SSL, which aligns pre-pooling local representations with the global image embedding, reaches 71.5% ImageNet-1K top-1 at 100 epochs and improves robustness over BYOL.

  4. Improving Pre-trained Self-Supervised Embeddings Through Effective Entropy Maximization

    cs.LG 2024-11 conditional novelty 5.0 of 10

    Continuing SSL pre-training for ten epochs with a criterion that maximizes one-dimensional marginal entropies and minimizes pairwise covariances improves downstream ImageNet classification, especially with few labels.

  5. An Empirical Study of Accuracy-Robustness Tradeoff and Training Efficiency in Self-Supervised Learning

    cs.CV 2025-01 conditional novelty 4.0 of 10

    CF-AMC-SSL, applying free adversarial training to 16-crop EMP-SSL, trains in 97 minutes versus 530 for robust EMP-SSL while improving PGD-8 robustness on CIFAR-10 from 28.49% to 33.34%.

Pith tools