Fast Inference in Sparse Coding Algorithms with Applications to Object Recognition
Koray Kavukcuoglu, Marc’Aurelio Ranzato, Yann LeCun · arXiv (Cornell University) · 2010
Adaptive sparse coding methods learn a possibly overcomplete set of basis functions, such that natural image patches can be reconstructed by linearly combining a small subset of these bases. The applicability of these methods to visual object recognition tasks has been limited because of the prohibitive cost of the optimization algorithms required to compute the sparse representation. In this work we propose a simple and efficient algorithm to learn basis functions. After training, this model also provides a fast and smooth approximator to the optimal representation, achieving even better accuracy than exact sparse coding algorithms on visual object recognition tasks.
Cited by 183 works. doi:10.48550/arxiv.1010.3467 Open-access PDF
Authors
References
- Gradient-based learning applied to document recognition
- Reducing the Dimensionality of Data with Neural Networks
- Training Products of Experts by Minimizing Contrastive Divergence
- Matching pursuits with time-frequency dictionaries
- Atomic Decomposition by Basis Pursuit
- Least angle regression
- Learning Generative Visual Models from Few Training Examples: An Incremental Bayesian Approach Tested on 101 Object Categories
- Sparse coding with an overcomplete basis set: A strategy employed by V1?
- Efficient Learning of Sparse Representations with an Energy-Based Model
- Optimally sparse representation in general (nonorthogonal) dictionaries via ℓ 1 minimization