Lossy Compression for Lossless Prediction · HackerPapers discussions

Lossy Compression for Lossless Prediction — mentioned by marauder777

Sun, 10 Apr 2022 23:06:13 GMT

While not exactly the idea in your post, this recent paper by Dubois, Bloem-Reddy, Ullrich, and Maddison on "Lossy Compression for Lossless Prediction" (https://arxiv.org/abs/2106.10800) kinda gets at this idea. From the abstract: Most data is automatically collected and only ever "seen" by algorithms. Yet, data compressors preserve perceptual fidelity rather than just the information needed by algorithms performing downstream tasks. In this paper, we characterize the bit-rate required to ensure high performance on all predictive tasks that are invariant under a set of transformations, such…

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

Lossy Compression for Lossless Prediction — mentioned by vpj

Wed, 23 Jun 2021 09:09:52 GMT

PDF: https://arxiv.org/pdf/2106.10800v1.pdf

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

Lossy Compression for Lossless Prediction — HN submission

Wed, 23 Jun 2021 02:04:27 GMT

Lossy Compression for Lossless Prediction

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