Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

count% of corpus
0.3%0.5%0.8%0.1% · Nov 201919922001200920172026
1 result for “HSQ”
Hyper-Sphere Quantization: Communication-Efficient SGD for Federated Learningcs.LG

HSQ reduces communication costs in federated learning.

problem High cost of communicating gradients in federated learning.
method Hyper-sphere quantization (HSQ) framework for efficient gradient compression.
result HSQ achieves O(log⁡d)O(\log d) per-iteration communication cost, significantly reducing costs without compromising accuracy.
federated learninggradient compressioncommunication efficiencyHSQlow communication cost
2019-11-12abs ↗pdf ↗