https://github.com/stanford-futuredata/sinkhorn-label-allocation

Sinkhorn Label Allocation is a label assignment method for semi-supervised self-training algorithms. The SLA algorithm is described in full in this ICML 2021 paper: https://arxiv.org/abs/2102.08622.
https://github.com/stanford-futuredata/sinkhorn-label-allocation

few-shot-learning image-classification optimal-transport semi-supervised-learning

Last synced: about 2 months ago
JSON representation

Sinkhorn Label Allocation is a label assignment method for semi-supervised self-training algorithms. The SLA algorithm is described in full in this ICML 2021 paper: https://arxiv.org/abs/2102.08622.

• Host: GitHub
• URL: https://github.com/stanford-futuredata/sinkhorn-label-allocation
• Owner: stanford-futuredata
• License: mit
• Created: 2021-02-16T08:02:43.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2021-06-15T05:08:35.000Z (over 3 years ago)
• Last Synced: 2024-07-10T17:02:16.470Z (2 months ago)
• Topics: few-shot-learning, image-classification, optimal-transport, semi-supervised-learning
• Language: Python
• Homepage:
• Size: 140 KB
• Stars: 53
• Watchers: 8
• Forks: 3
• Open Issues: 0

Awesome Lists containing this project

• my-awesome-starred - stanford-futuredata/sinkhorn-label-allocation - Sinkhorn Label Allocation is a label assignment method for semi-supervised self-training algorithms. The SLA algorithm is described in full in this ICML 2021 paper: https://arxiv.org/abs/2102.08622. (Python)