FeatherCast

AI is evolving rapidly, and much of the recent advancement is driven by Deep Learning, a machine learning technique inspired by the inner-working of the human brain. In this session, we will discuss what deep learning is, and the new capabilities it enables. We will dive into a few computer vision and natural language models that are demonstrating super-human performance, and to integrate these models into your existing Java system leveraging Apache MXNet – an open source deep learning framework – and it’s Java API. In our session, we will show a brief introduction to DL, specifically about applications using MXNet Gluon to do the training and Java to do inference with different DL models in Production systems. We would also demostrate how to run and deploy the DL models with MXNet Java API. By the end of the session, audience will learn how to leverage deep learning tasks using Apache MXNet, the various gotchas involved, and where/how to learn more.