[1Z3] [General Session] 2. Machine Learning

[1Z3-03] Disentangled Representation Learning From Sequential Data

〇Masanori Yamada¹, Heecheol Kim ³, Kosuke Miyoshi³, Hroshi Yamakawa^2,3 (1. NTT Secure Platform Laboratories, 2. The Whole Brain Architecture Initiative, 3. Dwango Artificial Intelligence Laboratory)

Keywords:Representation Learning, Sequential Data

We present a time convolutional variational ladder autoencoder (TCVLAE), which learns disentangled and interpretable representations from sequential data without supervision. For the simple 2d data set, the proposed model experimentally shows that it is possible to separate the meaning of time series data.