Deep learning neural networks have become a prominent tool in science and machine learning. This webinar will provide a brief overview of the main concepts of neural networks and feature discovery. We will also demonstrate the basic convolution neural network for digit recognition using TensorFlow and try to get an intuition about how the network performs and what defines deep learning in practice. We will also review some useful notes on running deep networks on HPCs, such as using tensorboard, notebooks, and batch jobs. 

Instructor
Paul Rodriguez
Computational Data Scientist, SDSC

Paul Rodriguez received his PhD in Cognitive Science at University of California, San Diego (UCSD) in 1999. He spent several years doing research in neural network modeling, dynamical systems simulations, time series analysis, and statistical methods for analysis and predictions in fMRI data. For the last several years he has been helping users from a variety of domains with performing big data analytics, applying machine learning models, implementing deep learning models, all in an HPC (High Performance Computing) environment.