Deep learning frameworks, such as convolutional neural networks (CNNs), have almost completely replaced other machine learning techniques for specific tasks such as image recognition using large training datasets. In this webinar, we will go over how CNNs, their training methods, and hardware evolved since LeNet first appeared in the late 1990's. We will examine the challenges that came along, and some key innovations that helped overcome these challenges. We will also look at a guide on how to get started with CNNs, some common pitfalls, and tips and tricks in training CNNs.
Professor Amit Sethi is a faculty member at Indian Institute of Technology Guwahati, and a visiting scholar at University of Illinois as Chicago. His current interests are computational pathology, deep learning, and non-negative matrix factorization. He also works in video classification and super resolution, and has previously worked in structure from motion and human visual perception. Before joining IIT Guwahati, he worked as a management consultant in ZS Associates' Chicago office, where he worked with big data in the healthcare sector. He obtained his PhD in ECE from University of Illinois at Urbana-Champaign, and BTech in EE from IIT Delhi.
Below is the video lecture:Written on March 11th, 2017 by