Stacked What-Where Auto-encoders
Explore this paper's citation graph
Summary
A novel architecture, the "stacked what-where auto-encoders" (SWWAE), which integrates discriminative and generative pathways and provides a unified approach to supervised, semi-supervised and unsupervised learning without relying on sampling during training.
- Type
- preprint
- Published
- 2015-06-08
- Cited by
- 263
- References
- 47
- Access
- Open access
- OpenAlex
- https://openalex.org/W630242894
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:7397342
Keywords
Discriminative model, Pooling, Computer science, ENCODE, Generative grammar
References
- Training Very Deep Networks
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- An Analysis of Unsupervised Pre-training in Light of Recent Advances
- Lateral Connections in Denoising Autoencoders Support Supervised Learning
- Learning to Linearize Under Uncertainty
- A Winner-Take-All Method for Training Sparse Convolutional Autoencoders
- Improving neural networks by preventing co-adaptation of feature detectors
- Classification using discriminative restricted Boltzmann machines
- A Sparse and Locally Shift Invariant Feature Extractor Applied to Document Images
- Semi-supervised Learning with Deep Generative Models
- Gradient-based learning applied to document recognition
- Multi-Task Bayesian Optimization
- Learning Fast Approximations of Sparse Coding
- Fractional Max-Pooling
- A Fast Learning Algorithm for Deep Belief Nets
- Unsupervised Learning of Invariant Feature Hierarchies with Applications to Object Recognition
- Learning invariant features through topographic filter maps
- 80 Million Tiny Images: A Large Data Set for Nonparametric Object and Scene Recognition
- Discriminative Unsupervised Feature Learning with Convolutional Neural Networks
- Deep learning via semi-supervised embedding
Cited by
- Deep spatial autoencoders for visuomotor learning
- Convolutional Clustering for Unsupervised Learning
- Revealing Fundamental Physics from the Daya Bay Neutrino Experiment Using Deep Neural Networks
- Walk and Learn: Facial Attribute Representation Learning from Egocentric Video and Contextual Data
- Learning Temporal Regularity in Video Sequences
- Autoconvolution for Unsupervised Feature Learning
- Augmenting Supervised Neural Networks with Unsupervised Objectives for Large-scale Image Classification
- Improved Techniques for Training GANs
- Improved Deep Learning of Object Category Using Pose Information
- Neural Photo Editing with Introspective Adversarial Networks
- Non-negative Autoencoder with Simplified Random Neural Network
- Deep unsupervised learning through spatial contrasting
- Semantic Noise Modeling for Better Representation Learning
- Adversarial Ladder Networks
- Spatial contrasting for deep unsupervised learning
- Semi-Supervised Learning with Context-Conditional Generative Adversarial Networks
- S3Pool: Pooling with Stochastic Spatial Sampling
- Disentangling factors of variation in deep representation using adversarial training
- Identifying and Categorizing Anomalies in Retinal Imaging Data
- Semi-Supervised Learning with the Deep Rendering Mixture Model
Related papers
- Progressive Stage-wise Learning for Unsupervised Feature Representation Enhancement
- Unsupervised Feature Learning and Deep Learning: A Review and New Perspectives
- On Minimal Variations for Unsupervised Representation Learning
- Geometry and Topology of Conceptual Representations of Simple Visual Data
- Learning Disentangled Representation by Exploiting Pretrained Generative Models: A Contrastive Learning View
- Different latent variables learning in variational autoencoder