We are rolling out a new series of events: a Reinforcement Learning Reading Group. We plan to host it weekly in Sandbox Suites on Wednesdays at 6:30PM.
One person with deep domain knowledge will present the paper, and then we will have a round table discussion.
This reading group will be covering recent advances in Reinforcement Learning and is not an introductory event. You will not benefit from this meetup if you don't have basic understanding of deep learning.
You will also need to understand basics of Reinforcement Learning, such as state, actions and policies, we will not explain them at the meetup. Here's a decent introduction to RL: http://neuro.cs.ut.ee/demystifying-deep-reinforcement-learning/
Reading the paper in advance is highly encouraged.
The first paper will be Time-Contrastive Networks: Self-Supervised Learning from Multi-View Observation (https://arxiv.org/abs/1704.06888). It will be presented by Pierre Sermanet who is the first author of the paper.