Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Data

Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Data
复制标题

DOI:
--
复制
发表时间:
2016-05
期刊:
ArXiv
影响因子:
--
通讯作者:
Maximilian Karl;Maximilian Sölch;Justin Bayer;Patrick van der Smagt
Maximilian Karl;Maximilian Sölch;Justin Bayer;Patrick van der Smagt
中科院分区:
其他
文献类型:
--
作者:
Maximilian Karl;Maximilian Sölch;Justin Bayer;Patrick van der Smagt

文献摘要

被引文献

相似文献

我们介绍了深度变分贝叶斯滤波器(DVBF),这是一种用于无监督学习和识别潜在马尔可夫状态空间模型的新方法。利用随机梯度变分贝叶斯的最新进展,DVBF可以通过变分推理克服难以处理的推理分布。因此,它可以处理具有时间和空间依赖性的高度非线性输入数据,例如没有领域知识的图像序列。我们的实验表明,通过转换启用反向传播强制执行状态空间假设,并显着提高潜在嵌入的信息内容。这也使得现实的长期预测。
We introduce Deep Variational Bayes Filters (DVBF), a new method for unsupervised learning and identification of latent Markovian state space models. Leveraging recent advances in Stochastic Gradient Variational Bayes, DVBF can overcome intractable inference distributions via variational inference. Thus, it can handle highly nonlinear input data with temporal and spatial dependencies such as image sequences without domain knowledge. Our experiments show that enabling backpropagation through transitions enforces state space assumptions and significantly improves information content of the latent embedding. This also enables realistic long-term prediction.