Deep Learning by NVIDIA
On Nov 28th, 2017, I attended NVIDIA Developer Connect Program in The Suryaa, New Delhi, India. They brought together AI developers, data scientists, researchers, academia, influencers and other stakeholders to look at AI trends, technology, use cases and India’s role and opportunities in what many are calling the fourth Industrial Revolution.
The attendees were shown the following car demo which embarked the right attention within us.
This was the first event I ever attended on the theme of Deep Learning. Mr. Sundara Ramalingam Nagalingam, head — Deep Learning Practice, NVIDIA Graphics Pvt Ltd. inaugurated the event with great enthusiasm and zeal. He talked about NVIDIA, about how deep learning is used in the research community and in industry to help solve many big data problems such as computer vision, speech recognition, and natural language processing.
Then came Mr. Vishal Dhupar, Managing Director, Asia South at NVIDIA. His keynote started with the following video. I would recommend you to go through it.
Towards its end, we were told that even the music of this video was machine built which gave us goosebumps. Throughout Mr. Dhupar’s talk he told of something which made us clap almost immediately every five minutes. Mr. Dhupar showed us the technologies used over the years starting from PC, Mobile Era, Cloud Computing to Artificial Intelligence. Over his talks, he talked of the NVIDIA Deep Learning SDK which provides high-performance tools and libraries to power innovative GPU-accelerated machine learning applications in the cloud, data centers, workstations, and embedded platforms. He showed us how these libraries can be used for deep learning primitives, inference, video analytics, linear algebra, sparse matrices, and multi-GPU communications. He also mentioned the highest GIT downloads for CUDA ( CUDA is NVIDIA’s parallel computing architecture that enables dramatic increases in computing performance by harnessing the power of the GPU (graphics processing unit)). The show stopper of his keynote was DGX-1 an integrated system for deep learning. It features eight Tesla P100 GPU accelerators connected through NVLink, the NVIDIA high-performance GPU interconnect, in a hybrid cube-mesh network.
He also talked about Robotaxi, their demand for extreme computing. He mentioned XAVIER, world’s first autonomous machine processor.
Then there were Dr. Lovekesh Vig, Sr. Scientist, TCS Research Labs who talked on Deep Learning and the future Enterprise. Dr. Santanu Choudhury, Director, CEERI who gave the light on Deep Learning Architectures for Image based Applications and also talked about student internships available at CEERI.
What I really enjoyed in the context of AI research in India was talk by academician Dr. Brejesh Lall, Asst. Prof., IIT Delhi who showcased his work on Driver assistance in Indian Traffic context. The closing note was from Nitendra Rajput, Senior Vice President, Analytics at Infoedge.
Another attraction of this event was demo labs by Boston which is showcased exactly in following clip.
Overall, the event intrigued the attendees to make a focus on deep learning. Unfortunately here in home ground we have just begun to see its worth but across the world it’s in the process of getting matured. Hence, we really need to gear up.
