Research map / Label-efficient and robust learning / Recognition and long-tail
Clustering by Maximizing Mutual Information Across Views
Kien Do, Truyen Tran, Svetha Venkatesh
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add the representation learning head and feature contrastive loss (ablation) | C-head only · CIFAR10 | ACC: 66.9 -> 79.9 | Improved |
| add the representation learning head and feature contrastive loss (ablation) | C-head only · STL10 | ACC: 61.2 -> 81.8 | Improved |
| add the representation learning head and feature contrastive loss (ablation) | C-head only · ImageNet-Dogs | ACC: 36.3 -> 46.1 | Improved |
| use log-of-dot-product critic instead of dot-product critic (ablation) | CRLC probability contrastive loss · CIFAR10 and CIFAR20 | ACC | Improved |
| use feature contrastive loss weight lambda2 of 10 instead of 0 (ablation) | CRLC · CIFAR20 | ACC | Improved |
| use too large feature contrastive weight (ablation) | CRLC · CIFAR20 | ACC | Hurt |
| use MemoryBank implementation instead of SimCLR (ablation) | CRLC · CIFAR10/20 | ACC | Mixed |
| use train and test sets for clustering on STL10 (ablation) | CRLC · STL10 | ACC: 57.6 -> 81.8 | Improved |
| use train and test sets for clustering on CIFAR20 (ablation) | CRLC · CIFAR20 | ACC: 43.4 -> 42.5 | No clear effect |
| use train and test sets for clustering on CIFAR10 (ablation) | CRLC · CIFAR10 | ACC: 79.4 -> 79.9 | No clear effect |
| train both heads in second stage instead of only the C-head (ablation) | two-stage CRLC · clustering | ACC | Hurt |
| increase the number of negative samples (ablation) | MemoryBank CRLC · CIFAR20 | ACC | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- ReMixMatch — ReMixMatch: Semi-Supervised Learning with Distribution Matching and Augmentation Anchoring (ICLR 2020) · better on 0 of 1 shared result
- MixMatch — MixMatch: A Holistic Approach to Semi-Supervised Learning (NeurIPS 2019) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Joint Unsupervised Learning of Deep Representations and Image Clusters (CVPR 2016 · similarity 0.90)
- Clustering-friendly Representation Learning via Instance Discrimination and Feature Decorrelation (ICLR 2021 · similarity 0.89)
- Deep Clustering via Joint Convolutional Autoencoder Embedding and Relative Entropy Minimization (ICCV 2017 · similarity 0.88)
- Jigsaw Clustering for Unsupervised Visual Representation Learning (CVPR 2021 · similarity 0.88)
- Learning Representation for Clustering Via Prototype Scattering and Positive Sampling (TPAMI 2022 · similarity 0.87)
- Image Clustering via the Principle of Rate Reduction in the Age of Pretrained Models (ICLR 2024 · similarity 0.87)
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.
