Sumários
MDPs and Reinforcement Learning
19 novembro 2013, 14:00 • Pedro Urbano Lima
The stochastic view od DES. Stochastic automata and Petri nets. Markov Decision Processes. Dynamic Programming solution of MDPs. Value Iteration.
Plan representation using Petri nets (2)
14 novembro 2013, 14:00 • Pedro Urbano Lima
Abstraction levels in Discrete Event Systems and different model views (untimed/logical, timed and stochastically timed). Plan qualitative analysis and PN problems: boundedness, conservation, liveness.