New opening for an internship position in the field of machine learning

Comparison of GPU and CPU MLP implementations for image classification.

This project aims at comparing the performance of GPU and CPU implementations of a state-of-the-art "deep" multi-layer perceptron. While the massive parallelism of GPU architectures allows important speed gains when using a batch learning procedure, it is not clear how this compares to stochastic gradient descent implemented in a more classical manner.

More info about the position can be found on our " Idiap online recruitment system"