DSpace Repository

q-VAE for Disentangled Representation Learning and Latent Dynamical Systems

Show simple item record

dc.contributor.author Kobayashi, Taisuke
dc.date.accessioned 2020-12-28T02:52:50Z
dc.date.available 2020-12-28T02:52:50Z
dc.date.issued 2020-07-17
dc.identifier.uri http://hdl.handle.net/10061/14175
dc.description.abstract A variational autoencoder (VAE) derived from Tsallis statistics called q-VAE is proposed. In the proposed method, a standard VAE is employed to statistically extract latent space hidden in sampled data, and this latent space helps make robots controllable in feasible computational time and cost. To improve the usefulness of the latent space, this letter focuses on disentangled representation learning, e.g., β-VAE, which is the baseline for it. Starting from a Tsallis statistics perspective, a new lower bound for the proposed q-VAE is derived to maximize the likelihood of the sampled data, which can be considered an adaptive β-VAE with deformed Kullback-Leibler divergence. To verify the benefits of the proposed q-VAE, a benchmark task to extract the latent space from the MNIST dataset was performed. The results demonstrate that the proposed q-VAE improved disentangled representation while maintaining the reconstruction accuracy of the data. In addition, it relaxes the independency condition between data, which is demonstrated by learning the latent dynamics of nonlinear dynamical systems. By combining disentangled representation, the proposed q-VAE achieves stable and accurate long-term state prediction from the initial state and the action sequence. ja_JP
dc.language.iso en ja_JP
dc.publisher IEEE ja_JP
dc.rights Copyright © 2020, IEEE ja_JP
dc.subject Feature extraction ja_JP
dc.subject Aerospace electronics ja_JP
dc.subject Standards ja_JP
dc.subject Robots ja_JP
dc.subject Machine learning ja_JP
dc.subject Decoding ja_JP
dc.subject Gaussian distribution ja_JP
dc.title q-VAE for Disentangled Representation Learning and Latent Dynamical Systems ja_JP
dc.type.nii Journal Article ja_JP
dc.contributor.transcription コバヤシ, タイスケ
dc.contributor.alternative 小林, 泰介
dc.textversion author ja_JP
dc.identifier.eissn 2377-3766
dc.identifier.jtitle IEEE Robotics and Automation Letters ( ja_JP
dc.identifier.volume 5 ja_JP
dc.identifier.issue 4 ja_JP
dc.identifier.spage 5669 ja_JP
dc.identifier.epage 5676 ja_JP
dc.relation.doi 10.1109/LRA.2020.3010206 ja_JP
dc.identifier.NAIST-ID 74653270 ja_JP


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Advanced Search

Browse

My Account