用人类反馈训练60亿参数时序模型，提升预测准确率33.21%
TimeHF: Billion-Scale Time Series Models Guided by Human Feedback
- 采用分块卷积嵌入与人类反馈优化机制构建大模型
- 在京东供应链中实现超2万商品自动补货，准确率提升33.21%
- 适合需要高精度时序预测的工业场景，如供应链管理
时序神经网络在实际应用中表现优异，但面临可扩展性差、泛化能力弱和零样本性能不佳等问题。受大语言模型启发，学界致力于构建大规模时序模型（LTM）以解决上述挑战。然而现有方法在训练复杂度、人类反馈适配和预测精度方面仍存困难。本文提出TimeHF，一种包含60亿参数的大规模时序模型构建新流程，引入分块卷积嵌入捕捉长时序特征，并设计名为时间序列策略优化（time-series policy optimization）的人类反馈机制。该模型部署于京东供应链系统，实现超过20,000种商品的自动化补货，相比现有方法预测准确率提升33.21%。本工作推动了大规模时序建模技术发展，并展现出显著的工业应用价值。
原文摘要 · Abstract (English)
Time series neural networks perform exceptionally well in real-world applications but encounter challenges such as limited scalability, poor generalization, and suboptimal zero-shot performance. Inspired by large language models, there is interest in developing large time series models (LTM) to address these issues. However, current methods struggle with training complexity, adapting human feedback, and achieving high predictive accuracy. We introduce TimeHF, a novel pipeline for creating LTMs with 6 billion parameters, incorporating human feedback. We use patch convolutional embedding to capture long time series information and design a human feedback mechanism called time-series policy optimization. Deployed in JD.com's supply chain, TimeHF handles automated replenishment for over 20,000 products, improving prediction accuracy by 33.21% over existing methods. This work advances LTM technology and shows significant industrial benefits.
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。