Unlocking Data Efficiency: The Compression Sweet Spot with Pareto Optimization by Arvind Sundararajan
dev.to·10h·
Discuss: DEV
📐Compression Mathematics
Generalised solutions and law of conservation of difficulty (2008)
terrytao.wordpress.com·8h·
Discuss: Hacker News
🎯Performance Proofs
A Generalized Information Bottleneck Theory of Deep Learning
arxiv.org·8h
⧗Information Bottleneck
Holosplit
atdotde.blogspot.com·1h·
Discuss: www.blogger.com
🔧KAITAI
3rd Largest Element: SIMD Edition
parallelprogrammer.substack.com·15h·
Discuss: Substack
⚡SIMD Optimization
The AI Trustworthiness Pareto: Balancing Compression, Accuracy, and Identifiability by Arvind Sundararajan
dev.to·16h·
Discuss: DEV
🧠Intelligence Compression
Scientists finally prove that a quantum computer can unconditionally outperform classical computers
phys.org·21h
⚛️Quantum Tape
Throwing Darts in Latent Space
guille.site·1d·
Discuss: Hacker News
🧠Machine Learning
Brownian motion: Modeling random fields over arbitrary geometries
stopthrowingrocks.github.io·1d·
Discuss: Lobsters, Hacker News
🕸️Tensor Networks
Adaptive Source-Channel Coding for Multi-User Semantic and Data Communications
arxiv.org·1d
🌊Distributed Codecs
Beyond ROC-AUC and KS: The Gini Coefficient, Explained Simply
towardsdatascience.com·21h
🎯Arithmetic Coding Theory
Time needed to factor large integers
johndcook.com·21h·
Discuss: Hacker News
🔱Galois Cryptography
Solving 2-SAT
nima101.github.io·9h·
Discuss: Hacker News
🔗Constraint Handling
Adaptive AI: Unleash the Power of Dynamic Block Decoding for Faster, Smarter LLMs
dev.to·8h·
Discuss: DEV
💻Local LLMs
A robust computational framework for the mixture-energy-consistent six-equation two-phase model with instantaneous mechanical relaxation terms
arxiv.org·8h
💧Liquid Types
Reducing Data Lake Storage by 52% with SciPy-Style Sparse Arrays in DuckDB
blog.sturdystatistics.com·23h·
Discuss: Hacker News
🦀Rusty Databases
A Puzzling Performance Puzzle
kodewerk.com·1d·
Discuss: Hacker News
🧪Binary Fuzzing
General Pruning Criteria for Fast SBL
arxiv.org·2d
⧗Information Bottleneck
Beyond Power Laws: Scaling Laws for Next-Token Prediction
francisbach.com·4d·
Discuss: Hacker News
🧮Kolmogorov Bounds
Unbiased Binning: Fairness-aware Attribute Representation
arxiv.org·2d
🧠Machine Learning