To Compress or Not to Compress- Self-Supervised Learning and Information Theory: A Review — mentioned by esafak

Wed, 20 Sep 2023 16:16:36 GMT

The article emphasizes the wrong thing, in my view. The interesting part is that compression -- without learning a model -- can be used for classification. This raises the question of what other information-theoretic measures can be used; cheaper, lossy ones. To Compress or Not to Compress- Self-Supervised Learning and Information Theory: A Review https://arxiv.org/abs/2304.09355\*

Paper and discussion history. A mention is not an endorsement.

To Compress or Not to Compress- Self-Supervised Learning and Information Theory: A Review — HN submission

Sat, 26 Aug 2023 14:17:55 GMT

To Compress or Not to Compress- Self-Supervised Learning and Information Theory

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

To Compress or Not to Compress- Self-Supervised Learning and Information Theory: A Review · HackerPapers discussions

To Compress or Not to Compress- Self-Supervised Learning and Information Theory: A Review — mentioned by esafak

To Compress or Not to Compress- Self-Supervised Learning and Information Theory: A Review — HN submission