Unsupervised Learning of Disentangled and Interpretable Representations from Sequential Data
Explore this paper's citation graph
Summary
A factorized hierarchical variational autoencoder, which learns disentangled and interpretable representations from sequential data without supervision by formulating it explicitly within a factorsized hierarchical graphical model that imposes sequence-dependent priors and sequence-independent priors to different sets of latent variables.
- Type
- preprint
- Published
- 2017-09-22
- Cited by
- 379
- References
- 47
- Access
- Open access
- OpenAlex
- https://openalex.org/W2758785877
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:39395448
Keywords
Prior probability, Computer science, Autoencoder, Sequence (biology), Artificial intelligence
References
- Support vector machines versus fast scoring in the low-dimensional total variability space for speaker verification
- A Recurrent Latent Variable Model for Sequential Data
- An Introduction to Variational Methods for Graphical Models
- The Kaldi Speech Recognition Toolkit
- DARPA TIMIT:: acoustic-phonetic continuous speech corpus CD-ROM, NIST speech disc 1-1.1
- Deep Convolutional Inverse Graphics Network
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- Gated Feedback Recurrent Neural Networks
- Auto-Encoding Variational Bayes
- Conditional restricted Boltzmann machine for voice conversion
- Hybrid speech recognition with Deep Bidirectional LSTM
- Long Short-Term Memory
- Semi-supervised Learning with Deep Generative Models
- Front-End Factor Analysis for Speaker Verification
- Pixel Recurrent Neural Networks
- Long short-term memory recurrent neural network architectures for large scale acoustic modeling
- An introduction to computational networks and the computational network toolkit (invited talk)
- A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues
- The Design for the Wall Street Journal-based CSR Corpus
- InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
Cited by
- Towards Bayesian Deep Learning: A Survey
- Unsupervised domain adaptation for robust speech recognition via variational autoencoder-based data augmentation
- Discovering Order in Unordered Datasets: Generative Markov Networks
- A Deep Generative Model for Disentangled Representations of Sequential Data
- Extracting Domain Invariant Features by Unsupervised Learning for Robust Automatic Speech Recognition
- Style Tokens: Unsupervised Style Modeling, Control and Transfer in End-to-End Speech Synthesis
- Towards Unsupervised Automatic Speech Recognition Trained by Unaligned Speech and Text only
- A Multi-Discriminator CycleGAN for Unsupervised Non-Parallel Speech Domain Adaptation
- Scalable Factorized Hierarchical Variational Autoencoder Training
- Disentangling Controllable and Uncontrollable Factors of Variation by Interacting with the World
- Discovering Interpretable Representations for Both Deep Generative and Discriminative Models
- Disentangling by Partitioning: A Representation Learning Framework for Multimodal Sensory Data
- Unsupervised Adaptation with Interpretable Disentangled Representations for Distant Conversational Speech Recognition
- InfoCatVAE: Representation Learning with Categorical Variational Autoencoders
- Augmented Cyclic Adversarial Learning for Domain Adaptation
- Deep appearance models for face rendering
- Phonetic-and-Semantic Embedding of Spoken words with Applications in Spoken Content Retrieval
- Deep Encoder-Decoder Models for Unsupervised Learning of Controllable Speech Synthesis
- Towards Learning Fine-Grained Disentangled Representations from Speech
- A Voice Conversion Framework with Tandem Feature Sparse Representation and Speaker-Adapted WaveNet Vocoder
Related papers
- A Comparative Evaluation of AutoEncoder-Based Unsupervised Anomaly Detection Methods Applied on Space Payload
- Performance Comparison of Three Types of Autoencoder Neural Networks
- The Learning Effect of Different Hidden Layers Stacked Autoencoder
- Combining an Autoencoder and a Variational Autoencoder for Explaining the Machine Learning Model Predictions
- The difference learning of hidden layer between autoencoder and variational autoencoder
- Assisted Diagnosis Research Based on Improved Deep Autoencoder
- Reconstruction of handwritten digit images using autoencoder neural networks
- A Stable Variational Autoencoder for Text Modelling
- FlowPrior: Learning Expressive Priors for Latent Variable Sentence Models