Sumários

Session L5

22 abril 2022, 15:30 • João Godinho Ribeiro

Reinforcement learning.

13. Exploration vs. exploitation

22 abril 2022, 09:30 • Francisco Melo

Exploration vs. exploitation (Chap. 9):

  • The prediction problem (9.1)
  • Prediction with complete information: weighted majority and exponentially weighted averager (9.2)
  • Adversarial bandits and EXP3 (9.4)
  • Stochastic multi-armed bandits and UCB (9.3)

  • Applications:

    • Monte Carlo tree search
    • TD-Gammon
    • Alpha-Go
Wrap-up.

13. Exploration vs. exploitation

21 abril 2022, 15:00 • Francisco Melo

Exploration vs. exploitation (Chap. 9):

  • The prediction problem (9.1)
  • Prediction with complete information: weighted majority and exponentially weighted averager (9.2)
  • Adversarial bandits and EXP3 (9.4)
  • Stochastic multi-armed bandits and UCB (9.3)

  • Applications:

    • Monte Carlo tree search
    • TD-Gammon
    • Alpha-Go
Wrap-up.

Session L5

21 abril 2022, 08:30 • Diogo Filipe de Sousa Carvalho

Reinforcement learning.

Session L5

20 abril 2022, 14:00 • José Alberto Rodrigues Pereira Sardinha

Reinforcement learning.