Yann LeCun, director of AI Research at Facebook and Silver Professor at New York University, appeared as guest speaker at NJIT’s Ying Wu College of Computing’s Distinguished Speaker Series Nov. 8.

Lecun’s presentation, titled “The Power and Limits of Deep Learning and AI: How Can Machines Learn as Efficiently as Animals and Humans?,” addressed the latest challenges and promising research shaping the groundbreaking field of deep learning — a field of machine learning based on computer algorithms that can learn and improve by themselves.

“Currently, a cat has more common sense than the smartest machine that we have,” LeCun said in front of a capacity audience at NJIT’s Central King Building. “Predictive learning is the main paradigm for learning that humans and animals seem to use, and we have not been able to reproduce in machines yet. When we learn how to reproduce it, we can then be able to have truly intelligent machines.”

LeCun discussed new approaches toward learning predictive models that can help reproduce human unsupervised predictive learning — or “common sense” — which may help impact technologies from self-driving cars and content filtering to medical image analysis. He concluded the discussion by fielding questions from NJIT students, faculty and guests.

Director of AI Research at Facebook Artificial Intelligence Research since 2013, LeCun has become best known for his work in deep learning and contribution to the development of the “convolutional neural network” method, which is widely used for image, video and speech recognition.