Rectified Linear Units Improve Restricted Boltzmann Machines
Explore this paper's citation graph
Summary
Restricted Boltzmann machines were developed using binary stochastic hidden units that learn features that are better for object recognition on the NORB dataset and face verification on the Labeled Faces in the Wild dataset.
- Type
- article
- Published
- 2010-06-21
- Cited by
- 18,989
- References
- 21
- OpenAlex
- https://openalex.org/W1665214252
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:15539264
Keywords
Boltzmann machine, Binary number, Sigmoid function, Computer science, Object (grammar)
References
- Diffusion Networks, Products of Experts, and Factor Analysis
- Labeled Faces in the Wild: A Database forStudying Face Recognition in Unconstrained Environments
- An empirical evaluation of deep architectures on problems with many factors of variation
- Phone recognition using Restricted Boltzmann Machines
- Restricted Boltzmann machines for collaborative filtering
- Replicated Softmax: an Undirected Topic Model
- Rate-coded Restricted Boltzmann Machines for Face Recognition
- Permitted and Forbidden Sets in Symmetric Threshold-Linear Networks
- Training Products of Experts by Minimizing Contrastive Divergence
- Learning methods for generic object recognition with invariance to pose and lighting
- A Fast Learning Algorithm for Deep Belief Nets
- Implicit Mixtures of Restricted Boltzmann Machines
- Learning a similarity metric discriminatively, with application to face verification
- Unsupervised learning of distributions on binary vectors using two layer networks
- Attribute and simile classifiers for face verification
- What is the best multi-stage architecture for object recognition?
- Scaling learning algorithms towards AI
- Reducing the Dimensionality of Data with Neural Networks
- A Hierarchical Community of Experts
- Similarity Scores Based on Background Samples
Cited by
- Regularization of Neural Networks using DropConnect
- Domain Adaptation for Large-Scale Sentiment Classification: A Deep Learning Approach
- Segregating Complex Sound Sources through Temporal Coherence
- Deep MRI brain extraction: A 3D convolutional neural network for skull stripping
- Noise Perturbation for Supervised Speech Separation
- Segmenting Retinal Blood Vessels With Deep Neural Networks
- dipIQ: Blind Image Quality Assessment by Learning-to-Rank Discriminable Image Pairs
- Quicksilver: Fast Predictive Image Registration – a Deep Learning Approach
- Mol2vec: Unsupervised Machine Learning Approach with Chemical Intuition
- Genome-wide prediction of cis-regulatory regions using supervised deep learning methods
- LeukocyteMask: An automated localization and segmentation method for leukocyte in blood smear images using deep neural networks
- Flow Field Reduction Via Reconstructing Vector Data From 3-D Streamlines Using Deep Learning
- A Deep Learning Reconstruction Framework for Differential Phase-Contrast Computed Tomography With Incomplete Data
- Artificial Intelligence-Based Smart Quality Inspection for Manufacturing
- A Deep Learning Pipeline for Image Understanding and Acoustic Modeling
- Occlusion Edge Detection in RGB-D Frames using Deep Convolutional Networks
- Unsupervised Feature Learning and Deep Learning: A Review and New Perspectives
- Learning Deep Representation for Face Alignment with Auxiliary Attributes
- Online Multi-Stage Deep Architectures for Feature Extraction and Object Recognition
- Dependency Parsing for Weibo: An Efficient Probabilistic Logic Programming Approach
Related papers
- Learning Multiple Layers of Features from Tiny Images
- Deep Learning
- Deep Residual Learning for Image Recognition
- ImageNet classification with deep convolutional neural networks
- ImageNet Large Scale Visual Recognition Challenge
- Gradient-based learning applied to document recognition
- ImageNet: A large-scale hierarchical image database