Choose timezone
Your profile timezone:
Abstract: Reinforcement learning algorithms have been able to achieve super-human performances in many scientific and technological applications. In this talk I will give an introduction to the the most successful algorithms and illustrate their advantages compared to classic supervised methods. I will then discuss possible future applications in high energy physics, which range from real time automatic control systems in accelerator facilities to handling the large combinatorial space in jet clustering algorithms.
Meeting ID: 912 1383 7955
Passcode: 069577