J. Meylahn

works on

algorithmic collusion 1 cooperation 1 epsilon-greedy 1 q-learning 1 reinforcement learning 1 repeated games 1

From the 1 of 5 linked papers with an AI index.

collaborators

Showing stat.MLShow all

1 paper · 1 filter

stat.ML2026

Beyond the Independence Assumption: Finite-Sample Guarantees for Deep Q-Learning under $I ¨ $ -Mixing

Leon Halgryn, Sophie Langer, Janusz M. Meylahn +1

Finite-sample analyses of deep Q-learning typically treat replayed data as independent, even though it is sampled from temporally dependent state-action trajectories. We study the…

Beyond the Independence Assumption: Finite-Sample Guarantees for Deep Q-Learning under I¨-Mixing

Beyond the Independence Assumption: Finite-Sample Guarantees for Deep Q-Learning under $I ¨ $ -Mixing