ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2501.12896
46
0

Irrational Complex Rotations Empower Low-bit Optimizers

22 January 2025
Zhen Tian
Wayne Xin Zhao
Zhicheng Dou
    MQ
ArXivPDFHTML
Abstract

In this paper, we propose a novel optimizer state compression algorithm, namely π\piπ-Quant, which leverages the properties of irrational numbers (e.g., π\piπ) for memory-efficient training. The core idea is based on our mathematical findings, which show that a pair of parameters can be represented by a single rotation angle using the complex rotation scheme. Building on this insight, we map the parameters into a complex space and perform quantization using the corresponding rotation angles. To efficiently integrate it into optimization process, we develop an efficient system of geometric equations that computes the precise rotation angles with linear complexity. We evaluate π\piπ-Quant on a wide range of tasks. Our experiments show that it can reduce the bit-width of parameters to 3.32-bit, achieving a 75% reduction in parameter scale and a 40% decrease in GPU memory usage, all while maintaining full accuracy.

View on arXiv
Comments on this paper