Learn-O-Matic: A Fully Automated Machine Learning Suite for Profile Retrieval Applications

This paper describes the Machine Learning suite Learn-O-Matic. Its key features are that it provides a completely automated framework for supervised learning with an easy-to-use web frontend which executes the complete learning process on NVIDIA based graphic cards. Meta parameters like the network architecture and the regularization term involved are optimised via state-of-the-art Reinforcement Learning techniques. The performance of the function approximator on the test set serves as the reward for the Reinforcement Learner. We show on data for ozone profile retrieval applications how to use Learn-O-Matic and provide results of the resulting retrieval system and of a wind power forecast system.

[1]  Volodymyr Mnih,et al.  CUDAMat: a CUDA-based matrix class for Python , 2009 .

[2]  John D. Owens,et al.  GPU Computing , 2008, Proceedings of the IEEE.

[3]  Radford M. Neal Pattern Recognition and Machine Learning , 2007, Technometrics.

[4]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[5]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[6]  Luca Maria Gambardella,et al.  Handwritten Digit Recognition with a Committee of Deep Neural Nets on GPUs , 2011, ArXiv.

[7]  Jürgen Schmidhuber,et al.  A committee of neural networks for traffic sign classification , 2011, The 2011 International Joint Conference on Neural Networks.

[8]  Frank Sehnke,et al.  Parameter-exploring policy gradients , 2010, Neural Networks.

[9]  Rajat Raina,et al.  Large-scale deep unsupervised learning using graphics processors , 2009, ICML '09.

[10]  A. Ng Feature selection, L1 vs. L2 regularization, and rotational invariance , 2004, Twenty-first international conference on Machine learning - ICML '04.