arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2303.04743cs.LGcs.AIstat.ML
收藏

基于双向先验模型的向量量化时间序列生成

Vector Quantized Time Series Generation with a Bidirectional Prior Model

Telenor Research(挪威电信研究院);BI Norwegian Business School(BI挪威商学院);Abelee

Daesoo Lee, Sara Malacarne, Erlend Aune

AI总结:

针对时间序列生成中GAN训练困难及RNN难以捕捉长程时间一致性的问题,提出TimeVQVAE,首次利用向量量化技术并在时频域分离高低频,结合双向Transformer学习离散先验,生成质量更优的时序信号。

摘要:

时间序列生成(TSG)研究主要集中在将生成对抗网络(GAN)与循环神经网络(RNN)变体结合使用。然而,训练GAN的根本局限性与挑战依然存在。此外,RNN系列通常难以处理相距较远时间步之间的时间一致性。受图像生成(IMG)领域成功的启发,我们提出了TimeVQVAE,据我们所知,这是首个使用向量量化(VQ)技术解决TSG问题的工作。此外,离散潜空间的先验是通过双向Transformer模型学习的,该模型能更好地捕捉全局时间一致性。我们还提出了在时频域中进行VQ建模,将其分离为低频(LF)和高频(HF)。这使我们能够保留时间序列的重要特征,进而生成质量更好的新合成信号,与竞争的TSG方法相比,在模块性上具有更尖锐的变化。我们的实验评估在UCR档案库的所有数据集上进行,使用了IMG文献中成熟的指标,如Fréchet inception distance和inception scores。我们的实现发布在GitHub上:https://github.com/ML4ITS/TimeVQVAE。

英文摘要:

Time series generation (TSG) studies have mainly focused on the use of Generative Adversarial Networks (GANs) combined with recurrent neural network (RNN) variants. However, the fundamental limitations and challenges of training GANs still remain. In addition, the RNN-family typically has difficulties with temporal consistency between distant timesteps. Motivated by the successes in the image generation (IMG) domain, we propose TimeVQVAE, the first work, to our knowledge, that uses vector quantization (VQ) techniques to address the TSG problem. Moreover, the priors of the discrete latent spaces are learned with bidirectional transformer models that can better capture global temporal consistency. We also propose VQ modeling in a time-frequency domain, separated into low-frequency (LF) and high-frequency (HF). This allows us to retain important characteristics of the time series and, in turn, generate new synthetic signals that are of better quality, with sharper changes in modularity, than its competing TSG methods. Our experimental evaluation is conducted on all datasets from the UCR archive, using well-established metrics in the IMG literature, such as Fréchet inception distance and inception scores. Our implementation on GitHub: \url{https://github.com/ML4ITS/TimeVQVAE}.

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。

↑