«  srpen  2022  »
Po Út St Čt So Ne
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31        

Zpět na kalendář

AIC Seminar: Aurélien Delage (INSA Lyon)

18.08.2022   14:00

We would like to invite you to a special summer seminar that will take place this Thursday, August 18 at 2:00 PM. Doctoral student Aurélien Delage from the National Institute of Applied Sciences of Lyon (INSA Lyon) is visiting the Department of Computer Science and will share his research in the field of Game Theory. You can find the full abstract below.

Heuristic Search Value Iteration can solve zero-sum Partially Observable Stochastic Games

State-of-the-art methods for solving 2-player zero-sum imperfect information games rely on linear programming or regret minimization, though not on dynamic programming (DP) or heuristic search (HS), while the latter are often at the core of state-of-the-art solvers for other sequential decision-making problems. In partially observable or collaborative settings (e.g., POMDPs and Dec-POMDPs), DP and HS require introducing an appropriate statistic that induces a fully observable problem as well as bounding (convex) approximators of the optimal value function. This approach has succeeded in some subclasses of 2-player zero-sum partially observable stochastic games (zs-POSGs) as well, but how to apply it in the general case still remains an open question. We answer it by (i) rigorously defining an equivalent game to work with which allows the application of DP and (ii) proving mathematical properties of the optimal value function that allow deriving bounds that come with solution strategies. We can thus design an HSVI-like solver that provably converges to an ε-optimal.

Místo konání
KN:E-205 (Karlovo náměstí 13, Praha 2)
Katedra počítačů FEL ČVUT
Kontaktní osoba
Karolína Poliaková, poliakar@fel.cvut.cz
Podrobnější informace