Standard low-precision formats are suboptimal for vector direction preservation compared to optimized scalar alphabets and spherical codes, with gaps quantified analytically and proven in Lean, and NVIDIA's E2M1 nearly matching the optimum for 4 bits.
A note on a method for generating points uniformly on n-dimensional spheres
3 Pith papers cite this work, alongside 387 external citations. Polarity classification is still indexing.
3
Pith papers citing it
387
external citations · external index
fields
cs.LG 3years
2026 3representative citing papers
EntroPath defines a free-energy dissimilarity from maximum-entropy random walk path ensembles and proves it converges to squared geodesic distance in the short-time limit via Varadhan's formula.
Thesis uses statistical mechanics to study DAM and RBM models for understanding memorization, low-dimensional learning, and adversarial robustness in neural networks.
citing papers explorer
-
Direction-Preserving Number Representations
Standard low-precision formats are suboptimal for vector direction preservation compared to optimized scalar alphabets and spherical codes, with gaps quantified analytically and proven in Lean, and NVIDIA's E2M1 nearly matching the optimum for 4 bits.
-
EntroPath: Maximum Entropy Path Ensemble Embedding for Manifold Learning
EntroPath defines a free-energy dissimilarity from maximum-entropy random walk path ensembles and proves it converges to squared geodesic distance in the short-time limit via Varadhan's formula.
-
Explaining Machine Learning and Memorization with Statistical Mechanics
Thesis uses statistical mechanics to study DAM and RBM models for understanding memorization, low-dimensional learning, and adversarial robustness in neural networks.