Deep Learning for Self-Driving Cars
We will provide an overview of how deep neural network based approaches can contribute to each individual component of an autonomous vehicle including scene perception, scene understanding, localization, mapping, control, planning, driver sensing, and the end-to-end driving task. We will discuss the strengths and limitations of the fundamental deep learning methods involved, including convolutional neural networks, recurrent neural networks, and policy networks for deep reinforcement learning in a complex, sparsely-supervised, safety-critical world.
Lex Fridman is a postdoc at MIT, working on computer vision and deep learning approaches in the context of self-driving cars with a human-in-the-loop. His work focuses on large-scale, real-world data, with the goal of building intelligent systems that have real world impact. Lex received his BS, MS, and PhD from Drexel University where he worked on applications of machine learning, computer vision, and decision fusion techniques in a number of fields including robotics, active authentication, activity recognition, and optimal resource allocation on multi-commodity networks. Before joining MIT, Lex was at Google working on machine learning and decision fusion methods for large-scale behavior-based authentication.