All seminars are held at the Department of Automatic Control, in the seminar room M:2112B on the second floor in the M-building, unless stated otherwise.
Seminar by Ather Gattami
Place: Seminar Room KC 3N27 at Dept. of Automatic Control
Contact: anders [dot] rantzer [at] control [dot] lth [dot] se
Save event to your calendar
The Department of Automatic Control is happy to announce a seminar by Ather Gattami.
When: Wednesday March 22, 10.30-11.30
Where: Seminar Room KC 3N27 at Automatic Control, Kemicentrum.
Speaker: Ather Gattami, AI Sweden, Stockholm
Title: Decentralized Online Bandit Optimization on Directed Graphs with Regret Bounds
We consider a decentralized multiplayer game, played over T rounds, with a leader-follower hierarchy described by a directed acyclic graph. For each round, the graph structure dictates the order of the players and how the players observe the actions of one another. By the end of each round, all players receive a joint bandit-reward based on their joint action that is used to update the player strategies towards the goal of minimizing the joint pseudo-regret. We present a learning algorithm inspired by the single-player multi-armed bandit problem and show that it achieves sub-linear joint pseudo-regret in the number of rounds for both adversarial and stochastic bandit rewards. Furthermore, we quantify the cost incurred due to the decentralized nature of our problem compared to the centralized setting.