Reinforcement learning algorithms for DASH video streaming

DASH is a video streaming standard developed in 2011; the servers have several copies of every video at different bitrates, leaving the clients complete freedom to adapt to the available bandwidth. This thesis presents two reinforcement learning algorithms to optimize this adaptation: Offline and Online. The Offline algorithm relies on a training phase to learn its environment, while the Online algorithm has a slimmer model and focuses on learning as quickly as possible, with no training phase