arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2009.12682cs.LGstat.ML
收藏

时间序列数据的决策感知条件GANs

Decision-Aware Conditional GANs for Time Series Data

Harvard University(哈佛大学);Massachusetts Institute of Technology(麻省理工学院)

He Sun, Zhun Deng, Hui Chen, David C. Parkes

摘要:

我们引入决策感知的时间序列条件生成对抗网络(DAT-CGAN)作为时间序列生成的方法。该框架采用了多Wasserstein损失结构化决策相关的数量,捕捉决策相关的数据的异质性,并提供新的有效性,在支持最终用户的决策过程。我们通过重叠块采样方法提高了样本效率,并从理论上描述了DAT-CGAN的泛化性能。金融时间序列的多时间步投资组合选择问题的框架证明。我们在基础数据和不同的决策相关量方面表现出比基于GAN的强大基线更好的生成质量。

英文摘要:

We introduce the decision-aware time-series conditional generative adversarial network (DAT-CGAN) as a method for time-series generation. The framework adopts a multi-Wasserstein loss on structured decision-related quantities, capturing the heterogeneity of decision-related data and providing new effectiveness in supporting the decision processes of end users. We improve sample efficiency through an overlapped block-sampling method, and provide a theoretical characterization of the generalization properties of DAT-CGAN. The framework is demonstrated on financial time series for a multi-time-step portfolio choice problem. We demonstrate better generative quality in regard to underlying data and different decision-related quantities than strong, GAN-based baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑