Large-scale deep unsupervised learning using graphics processors
Rajat Raina, Anand Madhavan, Andrew Ng · International Conference on Machine Learning · 2009
The promise of unsupervised learning methods lies in their potential to use vast amounts of unlabeled data to learn complex, highly nonlinear models with millions of free parameters. We consider two well-known unsupervised learning models, deep belief networks (DBNs) and sparse coding, that have recently been applied to a flurry of machine learning applications (Hinton & Salakhutdinov, 2006; Raina et al., 2007). Unfortunately, current learning algorithms for both models are too slow for large-scale applications, forcing researchers to focus on smaller-scale models, or to use fewer training examples.
Cited by 731 works. doi:10.1145/1553374.1553486
Authors
References
- A Fast Learning Algorithm for Deep Belief Nets
- Reducing the Dimensionality of Data with Neural Networks
- Regression Shrinkage and Selection Via the Lasso
- MapReduce
- Training Products of Experts by Minimizing Contrastive Divergence
- Greedy Layer-Wise Training of Deep Networks
- Least angle regression
- Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations
- Emergence of simple-cell receptive field properties by learning a sparse code for natural images
- Efficient sparse coding algorithms