跳至正文
返回文库全部文档

加密货币预测模型的前向门控替换

文章 arXiv papers · 作者: Aditya Dutta
阅读原文

总结

论文为定期重新训练的预测系统提出一种部署策略：继续运行现有模型，同时让经过预热重拟合的候选模型在服务链路之外运行；随后在同一个后续周的延迟标签上比较两者。只有当候选模型的负对数似然达到固定的配对优势门槛时，才会替换现有模型。该策略旨在避免按日历自动切换模型，因为新拟合的模型可能不如持续学习的现有模型。

在 Binance 两个独立时期的历史回放中，与按日历替换、自动晋升和持续维护相比，该策略降低了负对数似然。它晋升的候选模型更少，也减少了已部署模型的变更次数。作者报告称，在不同随机种子、试验预算、晋升幅度、较早的资产面板以及监督学习目标下，结果方向一致。这些是针对指定加密货币永续期货合约的回放结果；现有描述无法证明其在实盘部署或其他市场中的表现。

核心观点

  • 该策略在替换现有模型前，会将经过预热重拟合的候选模型与持续运行的模型一同评估。
  • 两个模型都在相同的下一周延迟标签上接受评估。
  • 只有负对数似然达到固定的配对改进门槛，候选模型才会晋升。
  • 历史回放报告称，与列出的对比策略相比，预测损失更低，已部署模型的变更也更少。
  • 报告的稳健性检查涵盖随机种子、晋升设置、资产面板和不同的监督学习目标。

标签

加密资产机器学习回测永续合约
全文
# Train Often, Deploy Selectively: Forward-Gated Model Replacement in Crypto Markets


# Train Often, Deploy Selectively: Forward-Gated Model Replacement in Crypto Markets









Production forecasting systems retrain models regularly, but a retrained candidate does not necessarily outperform a continuously maintained incumbent that has continued to learn. We introduce Shadow Before Swap (SBS), a deployment policy that warm-refits a challenger off the serving path, evaluates it against the maintained incumbent on the same next week of delayed labels, and promotes it only after a fixed paired negative-log-likelihood (NLL) advantage. In historical replay over two nonoverlapping Binance episodes spanning 48 UTC weeks, three seeds, eight underlyings, and two perpetual-futures contract types, SBS reduces NLL by 0.1472% relative to calendar replacement, 0.0755% relative to schedule-matched automatic promotion, and 0.0428% relative to continuous maintenance. The corresponding episode-stratified four-week block intervals are 0.1139%-0.1754%, 0.0521%-0.0980%, and 0.0301%-0.0554%, respectively. SBS promotes 114 of 528 challengers, reducing deployed model changes by 78.4% while improving the serving trajectory. The effect remains directionally consistent across seeds, trial budgets, promotion margins, an earlier 20-asset panel, and a topology-matched supervised objective. SBS thus provides a practical deployment policy that improves probabilistic forecasts while limiting consequential model-state transitions.

在遵守原作品许可的前提下，附作者信息全文展示。 许可协议: abstract CC0

此摘要由 Stratmill 研究智能体根据原文撰写，并非原文副本。