Dear students,
we approach the end of the teaching term. Today (May 15) and again in two weeks (May 29) we cover Deep Reinforcement Learning. You can view Deep Reinforcement Learning as the ultimate outcome of this class, since it uses the algorithms of RL (covered in the first few weeks of the class) and combines these with Deep Networks, trained by BackProp (covered in the last few weeks). Today is more conceptual, but in two weeks we turn to 'real' applications.
As announced in the fist week, there is no class on May 22 (Friday after Ascension). As a compensation, each week the lectures have been longer than the customary 90 minutes. So if you wondered why I upload sometimes 110 min worth of video time, this is why.
I hope you will enjoy the lectures during the remaining weeks.
Wulfram GERSTNER
