Pieter Abbeel

• deep reinforcement learning
• reinforcement leraning
• autonomous helicopter flight
• ImageNet (Geoffrey Hinton, Toronto team)
• AlexNet
• Supervise learning is about learning an input and output mapping
• Reinforcement learning: where does the data even come from? When you have data, how do you do credit assignment?
• andrej karpathy (Sr director at Tesla) deep learning videos
• Berkeley Deep Reinforcement Learning online class http://rail.eecs.berkeley.edu/deeprlcourse/
• Not just read things or watch videos, but try things out with frameworks like TensorFlow
• Behavioral cloning just supervise learning to mimic whatever the person is doing and then gradually later on, the reinforcement learning to have it think about longer time horizons

<end>