Can We Learn to Learn
A key challenge in the current state of machine intelligence is its inability to generalize and adapt to new tasks. Humans are able to adapt intelligently because of our fundamental brain structure that learns how to learn. The idea of teaching machines to learn how to learn, known as meta-learning, has been shown to be a promising approach in enabling agents to adapt to new tasks. In this talk, Bonnie will present the results of meta reinforcement learning applied to both research and application domains, as well as the development and the frontiers of this approach.
Bonnie Li is a Machine Learning researcher who is passionate about pushing the current boundaries of the field. At 17 year old, Bonnie is working on fundamental research in Reinforcement Learning at Mila under Yoshua Bengio. Bonnie holds a Deep Reinforcement Learning nanodegree from Udacity and was mentored by Microsoft. She is currently working on meta learning for efficient exploration, through which she hopes will lead us closer to artificial general intelligence.