March 11-12, 2019: Seminar by Dr. Saurav Basu, IBM India Research Laboratory

Deep Learning: Recent Interests in Theory and Practice


Dr. Saurav Basu

Research Staff Member

IBM India Research Laboratory

New Delhi, India


Time: 11:30 AM to 1:00 PM

Time: 2:00 PM to 3:00 PM

11th March 2019

Introduction to Deep Learning

Deep Learning Analysis

12th March 2019

Industry Application of Deep Learning


Venue: ECSU Seminar Room, 9th Floor, S. N. Bose Bhavan (Library Building)


Abstract: Deep learning has admittedly been a breakthrough in the recent history of AI, and vigorous research is directed towards building progressively complex networks that can train larger and more unstructured data such as video and text. A flurry of activity abounds in universities attempting to uncover fundamental characteristics of depth in learning. Industry hopes to reshape the future data-based action by investing heavily on compute infrastructure and building deep learning frameworks. At such a crucial juncture for deep learning and artificial intelligence in general, we take a look at the state of the art in deep learning, their motivation and applicability.

This will be a three part talk. It will start with the motivation and basics of neural nets, and recent modifications and techniques that has re-branded neural nets into deep learning in the first talk. We will cover the fundamentals and look at best practices, common architectures and popular frameworks. The second part of the talk will focus on some interesting theoretical problems that has been attracting the attention of deep learning researchers and industry practitioners. We will outline a few of them and also what IBM is contributing in this space. The last part of the talk will focus on applied aspects and current industrial topics in deep learning. We will touch upon some popular deep learning initiatives in industry and comment on the future of the industry.

Brief Biography: Dr. Saurav Basu has been working as a Research Scientist in IBM Research, Delhi since 2013. In his tenure at IBM, he has worked on simulation based optimization for scientific computing, optimization for discrete event simulation and machine learning based data analysis. His current focus is on theoretical and practical aspects of deep learning. Specifically, he is looking at understanding the fundamental contribution of depth in deep learning from a theoretical perspective, and scalability and convergence of industrial scale deep learning of massive datasets from an applied perspective. Earlier, has had finished his B.E from Jadavpur University, M.Tech in Computer Science from ISI Kolkata, and PhD in Electrical Engineering from University of Virginia. He has also worked as a Postdoctoral Fellow in the Carnegie Mellon University where he had researched the application of machine learning on image analysis.

Apply here:


Last date for applying online 9th March 2019

Notification to selected candidates 10th March 2019

For queries, please contact: /