NVIDIA 推出 GPU 加速金融工具聚类工作流 AdaptGrow
原标题:GPU-Accelerated Clustering for Financial Instruments at Scale
AI 摘要
NVIDIA 技术博客介绍了一种基于 GPU 加速的金融工具聚类工作流,使用 AdaptGrow 算法处理滚动相关性和尾部依赖矩阵,支持硬聚类、软因子载荷和结构突变信号。该工作流通过内存高效的 SymNMF 公式,使约 10 万工具可在单个 GB200 GPU 上处理,并支持跨 16 节点扩展至百万工具。实验显示,在 10 万工具规模下,AdaptGrow 在相关性和 TPDM 上分别于约 13 秒和 12.4 秒内收敛,展示了显著的性能提升。
正文节选
Use AdaptGrow, a GPU-accelerated matrix factorization algorithm, to turn rolling correlation and tail-dependence matrices into hard clusters, soft factor loadings, and structural-break signals at single-GPU and multi-node scale Quant strategies routinely group instruments for portfolio construction, risk aggregation, statistical arbitrage, and trade surveillance. Incorrect groupings can make concentrated positions appear diversified, obscure risk shared across nominal boundaries, and select stat