An empirical evaluation of deep architectures on problems with many factors of variation

@inproceedings{Larochelle2007AnEE,
  title={An empirical evaluation of deep architectures on problems with many factors of variation},
  author={H. Larochelle and D. Erhan and Aaron C. Courville and J. Bergstra and Yoshua Bengio},
  booktitle={ICML '07},
  year={2007}
}
Recently, several learning algorithms relying on models with deep architectures have been proposed. Though they have demonstrated impressive performance, to date, they have only been evaluated on relatively simple problems such as digit recognition in a controlled environment, for which many machine learning algorithms already report reasonable results. Here, we present a series of experiments which indicate that these models show promise in solving harder learning problems that exhibit many… Expand
866 Citations
An Algorithm for Training Polynomial Networks
  • 30
  • Highly Influenced
  • PDF
To go deep or wide in learning?
  • 18
  • PDF
On the Expressive Power of Deep Architectures
  • 14
  • PDF
On the Expressive Power of Deep Architectures
  • 216
  • PDF
Deep Learners Benefit More from Out-of-Distribution Examples
  • 99
  • PDF
Visualizing Higher-Layer Features of a Deep Network
  • 771
  • PDF
A Framework for Selecting Deep Learning Hyper-parameters
  • 7
  • PDF
Deep, super-narrow neural network is a universal classifier
  • L. Szymanski, B. McCane
  • Computer Science
  • The 2012 International Joint Conference on Neural Networks (IJCNN)
  • 2012
  • 4
...
1
2
3
4
5
...

References

SHOWING 1-3 OF 3 REFERENCES
A Fast Learning Algorithm for Deep Belief Nets
  • 11,871
  • Highly Influential
  • PDF
Learning a Nonlinear Embedding by Preserving Class Neighbourhood Structure
  • 440
  • Highly Influential
  • PDF
Learning methods for generic object recognition with invariance to pose and lighting
  • Y. LeCun, F. Huang, L. Bottou
  • Computer Science
  • Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.
  • 2004
  • 1,196
  • Highly Influential
  • PDF