Predictive variational autoencoder for learning robust representations of time-series data
Predictive variational autoencoder for learning robust representations of time-series data
复制标题
DOI：
10.48550/arxiv.2312.06932
复制
发表时间：
2023-12
期刊：
影响因子：
--
通讯作者：
Julia Huiming Wang;Dexter Tsin;Tatiana Engel
中科院分区：
文献类型：
--
作者：
Julia Huiming Wang;Dexter Tsin;Tatiana Engel
变异自动编码器(VAEs)已被广泛应用于发现控制神经活动和动物行为的低维潜在因素。然而，如果没有仔细的模型选择，未被覆盖的潜在因素可能会反映数据中的噪音，而不是真正的潜在特征，从而使这种表示不适合科学解释。这个问题的现有解决方案包括引入特定于特定数据类型的额外测量变量或数据扩充。我们提出了一种预测下一个时间点的VAE体系结构，并表明它减轻了虚假特征的学习。此外，我们还在潜在空间中引入了一种基于时间光滑度的模型选择度量。我们证明了这两个约束对VAE是平滑的，随着时间的推移产生了稳健的潜在表示，并忠实地恢复了合成数据集上的潜在因素。
Variational autoencoders (VAEs) have been used extensively to discover low-dimensional latent factors governing neural activity and animal behavior. However, without careful model selection, the uncovered latent factors may reflect noise in the data rather than true underlying features, rendering such representations unsuitable for scientific interpretation. Existing solutions to this problem involve introducing additional measured variables or data augmentations specific to a particular data type. We propose a VAE architecture that predicts the next point in time and show that it mitigates the learning of spurious features. In addition, we introduce a model selection metric based on smoothness over time in the latent space. We show that together these two constraints on VAEs to be smooth over time produce robust latent representations and faithfully recover latent factors on synthetic datasets.