[Conference] ICML 2023 Highlights

ICML 2023 Highlights!

지난 7월 23일부터 7월 29일까지 ICML 2023이 진행되었습니다! 그 중에서 관심을 가지고 볼만한 paper들을 요약한 포스트들을 공유드리려고 합니다.

1. Arithmetic Sampling: Parallel Diverse Decoding for Large Language Models
   • Notion Link
2. RankMe: Assessing the Downstream Performance of Pretrained Self-Supervised Representations by Their Rank
   • Notion Link
3. Evaluating Self-Supervised Learning via Risk Decomposition
   • Notion Link
4. Learning-Rate-Free Learning by D-Adaptation
   • Notion Link
5. Which Features are Learned by Contrastive Learning? On the Role of Simplicity Bias in Class Collapse and Feature Suppression
   • Notion Link
6. Tutorial : ML Theory
   • Notion Link