AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

Cluster-aware Semi-supervised Learning: Relational Knowledge Distillation Provably Learns Clustering

Yijun Dong, Kevin Miller, Qi Lei, Rachel Ward

NeurIPS 2023 · Label-efficient and robust learning › Knowledge distillation · approach: Patient consistent distillation · 1 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
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-10, Uniform labels, N=50000, High augmentationtop-1 accuracy: 89.47 -> 89.88No clear effect
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-10, Uniform labels, N=50000, Weak augmentationtop-1 accuracy: 81.36 -> 84.61Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-10, Uniform labels, N=25000, High augmentationtop-1 accuracy: 81.73 -> 87.06Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-10, StochasticGreedy labels, N=50000, High augmentationtop-1 accuracy: 85.48 -> 91.19Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-10, StochasticGreedy labels, N=40000, Weak augmentationtop-1 accuracy: 70.70 -> 85.12Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-10, StochasticGreedy labels, N=25000, Medium augmentationtop-1 accuracy: 73.97 -> 82.09Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-100, Uniform labels, N=50000, High augmentationtop-1 accuracy: 46.85 -> 49.33Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-100, Uniform labels, N=40000, Medium augmentationtop-1 accuracy: 42.98 -> 45.78Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-100, Uniform labels, N=25000, Weak augmentationtop-1 accuracy: 33.59 -> 35.69Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-100, StochasticGreedy labels, N=50000, Medium augmentationtop-1 accuracy: 48.64 -> 50.67Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-100, StochasticGreedy labels, N=40000, High augmentationtop-1 accuracy: 45.93 -> 47.62Improved
add relational knowledge distillation (RKD) loss to FixMatchFixMatch · CIFAR-100, StochasticGreedy labels, N=25000, Weak augmentationtop-1 accuracy: 38.22 -> 39.20Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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

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