Hyderabad: Meta's Chief AI Scientist and Vice-President Yann LeCun delivered a lecture at the Indian Institute of Technology Madras where he talked about the future of AI and addressed the topic-- How could machines reach human-level intelligence?
Dr Yann LeCun, Vice-President and Chief AI Scientist at Meta, highlighted that current AI architectures lack essential human-like intelligence traits such as common sense and complex reasoning. He introduced a cognitive architecture featuring a predictive world model, which aims to advance AI towards human-level capabilities by predicting action outcomes and planning sequences to meet objectives, using a self-supervised learning approach called Joint Embedding Predictive Architecture (JEPA).
He explained that the JEPA learns to encode percepts and predict future percepts, enhancing image and video understanding. It can detect unphysical events in videos and perform planning by predicting action sequences to achieve target states.
A faculty at New York University, Dr Yann LeCun was awarded the 2018 Turing Award, which is often referred to as the 'Nobel Prize in Computer Science' for his work on Deep Learning. IIT Madras launched the ‘Prof. Subra Suresh Institute Lecture Series’ during January 2020 in honour of one of the most distinguished alumni Prof Subra Suresh. It brings leading international scholars, engineers, and scientists to IIT Madras on a periodic basis, up to twice every year, to discuss key advances in scientific discoveries and technological innovation.
Prof Subra Suresh has held prestigious roles including Director of the US National Science Foundation, Dean of Engineering at MIT, and President of Carnegie Mellon University. He is notably the first IIT alumnus and Indian-born academic to achieve these leadership positions.
