Abstract
We transformed the dynamic routing and spectrum allocation problem into an Arcade-like game and trained a reinforcement learning agent to play it. Preliminary results show the agent outperforms the K-SP-FF heuristic.
© 2023 The Author(s)
PDF Article | Presentation VideoMore Like This
P. Chanclou, B. Capelle, B. Charbonnier, J.-L. Courant, Y. Denis, N. Genay, S. Gosselin, D. Kurz, B. Landousies, E. Le Bris, B. Le Guyader, A. Pizzinat, and F. Saliou
NTu2J.1 National Fiber Optic Engineers Conference (NFOEC) 2013
Qian Wang, H. Shakespear-Miles, Xiaoliang Chen, Marc Ruiz, Zhaohui Li, and Luis Velasco
Tu2D.4 Optical Fiber Communication Conference (OFC) 2023
C. Hernández-Chulde, R. Casellas, R. Martínez, R. Vilalta, and R. Muñoz
Tu3D.7 Optical Fiber Communication Conference (OFC) 2023