JAX中实现自缩放拟牛顿法，方便高效优化
Self-Scaled Broyden Family of Quasi-Newton Methods in JAX
- 基于JAX和Optimistix库，集成多种拟牛顿算法
- 支持强Wolfe条件的Zoom线搜索，确保收敛性
- 适合需要稳定优化的机器学习研究者使用
我们提供了自缩放拟牛顿法族在JAX中的完整实现，与JAX完全兼容，并基于Optimistix优化库构建。实现包括BFGS、DFP、Broyden及其自缩放变体（SSBFGS、SSDFP、SSBroyden），并集成满足强Wolfe条件的Zoom线搜索。本文为简短技术说明，非研究论文，不声称新贡献，旨在记录实现细节，促进这些优化器在JAX社区中的应用。代码已公开于https://github.com/IvanBioli/ssbroyden_optimistix.git。
原文摘要 · Abstract (English)
We present a JAX implementation of the Self-Scaled Broyden family of quasi-Newton methods, fully compatible with JAX and building on the Optimistix~\cite{rader_optimistix_2024} optimisation library. The implementation includes BFGS, DFP, Broyden and their Self-Scaled variants(SSBFGS, SSDFP, SSBroyden), together with a Zoom line search satisfying the strong Wolfe conditions. This is a short technical note, not a research paper, as it does not claim any novel contribution; its purpose is to document the implementation and ease the adoption of these optimisers within the JAX community. The code is available at https://github.com/IvanBioli/ssbroyden_optimistix.git.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。