AI Webinar Series on Deep Learning

The "AI Webinar Series on Deep Learning" will be held from June 29 to July 2 (detailed agenda follows). The series is promoted by the Italian NVIDIA AI Technology Centre (NVAITC) and aims at exploring the fundamentals of deep learning from building and training neural networks, optimizing data loading and performance through mixed-precision and parallelization, to deploying trained model in production for inference. Attendees will learn how to design, train, optimize, profile and deploy a deep neural network using NVIDIA technologies. The webinars are meant for PhD students, postdoc, researchers working with Deep Neural Networks.

• Session 0: Linear Regression in Pytorch (20’ + 20’ q/a) - Christian Hundt
• Session 1: Convolutional Neural Networks (20’ + 20’ q/a) - Christian Hundt

• Session 2: Efficient Data Loading using DALI (20’ + 20’ q/a) - Giuseppe Fiameni
• Session 3: Mixed Precision Training using Apex (20’ + 20’ q/a) - Paul Graham

• Session 4: Multi-GPU Training using Horovod (20’ + 20’ q/a) - Gunter Roeth
• Session 5: Deploying Models with TensorRT (20’ + 20’ q/a) - Niki Loppi
• Session 6: Profiling with NVTX (20’ + 20’ q/a) - Giuseppe Fiameni

NVIDIA Webex