Research map / Probabilistic and scientific modeling / Molecules and proteins
The Information Sieve
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| use 20 sieve representatives instead of 20 random representatives | single-layer sieve · Permuted binarized MNIST compression | bits per digit: 293 -> 266 | Helped |
| use 50 sieve representatives instead of 50 random representatives | single-layer sieve · Permuted binarized MNIST compression | bits per digit: 279 -> 252 | Helped |
| use 100 sieve representatives instead of 100 random representatives | single-layer sieve · Permuted binarized MNIST compression | bits per digit: 267 -> 243 | Helped |
| increase number of sieve representatives from 20 to 100 (ablation) | single-layer sieve · Permuted binarized MNIST compression | bits per digit: 266 -> 243 | Helped |
| increase number of random representatives from 20 to 100 (ablation) | random representatives baseline · Permuted binarized MNIST compression | bits per digit: 293 -> 267 | Helped |
| replace gzip with sieve 100 representatives | compression pipeline · Permuted binarized MNIST compression | bits per digit: 328 -> 243 | Helped |
| replace bitwise per-pixel coding with 20 sieve representatives | compression · Permuted binarized MNIST compression | bits per digit: 297 -> 266 | Helped |
| replace Huffman coding with bitwise Shannon coding | compression · Permuted binarized MNIST compression | bits per digit: 376 -> 297 | Helped |
Closest papers by meaning
- Learning Discrete Concepts in Latent Hierarchical Models (NeurIPS 2024 · similarity 0.85)
- On the Provable Advantage of Unsupervised Pretraining (ICLR 2024 · similarity 0.85)
- I-Con: A Unifying Framework for Representation Learning (ICLR 2025 · similarity 0.85)
- Connecting Jensen–Shannon and Kullback–Leibler Divergences: A New Bound for Representation Learning (NeurIPS 2025 · similarity 0.85)
- Generalization Guarantees for Representation Learning via Data-Dependent Gaussian Mixture Priors (ICLR 2025 · similarity 0.85)
- Cross-Entropy Is All You Need To Invert the Data Generating Process (ICLR 2025 · similarity 0.85)
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.