Research map / Architectures and efficiency / Quantization
Angular Networks: Low-Bit Learning from Randomized Similarity Estimators
Ali Ahmed, Saeid Pourmand, Muhammad Awais, Rehan Farooq, Alireza Aghasi
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes Angular Networks, which replace Euclidean linear layers with Angular Layers built on quantized random-feature estimators of cosine similarity for low-bit learning.
Problem
Low-precision neural networks degrade because they approximate Euclidean linear operations under limited precision.
Approach
Reinterprets neural computation in terms of directional (angular) similarity and constructs quantized random-feature estimators of cosine interactions, replacing standard linear layers with Angular Layers.
Closest papers by meaning
- Linear Symmetric Quantization of Neural Networks for Low-precision Integer Hardware (ICLR 2020 · similarity 0.86)
- Spherization Layer: Representation Using Only Angles (NeurIPS 2022 · similarity 0.86)
- Training invariances and the low-rank phenomenon: beyond linear networks (ICLR 2022 · similarity 0.86)
- Network Quantization With Element-Wise Gradient Scaling (CVPR 2021 · similarity 0.86)
- On Making $SE(2)$-Invariant Networks Optimal (NeurIPS 2026 · similarity 0.84)
- Asymptotics of Learning with Deep Structured (Random) Features (ICML 2024 · similarity 0.84)
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.
