Many recent advances in computer vision, speech recognition and Natural Language Processing (NLP) have come from Deep Learning research. One key enabler has been the use of GPUs to train bigger models on larger datasets. The Python module "Theano" can generate GPU code automatically - so that high level ideas can be implemented in high performance code seamlessly within the Python ecosystem.
The following topics will be covered, though not necessarily in the same order as below, since some 'Deep Learning' motivation might be good to introduce earlier on.
