Reinforcement-Learning-based Network Design and Control with Stepwise Reward Variation and Link-Adjacency Embedding

被引：0

作者：

Cruzado, Kenji ^{[1
]}

Shiraki, Ryuta ^{[1
]}

Mori, Yojiro ^{[1
]}

Tanaka, Takafumi ^{[2
]}

Higashimori, Katsuaki ^{[2
]}

Inuzuka, Fumikazu ^{[2
]}

Ohara, Takuya ^{[2
]}

Hasegawa, Hiroshi ^{[1
]}

机构：

[1] Nagoya University, Furo-cho, Chikusa, Aichi, Nagoya,464-8603, Japan

[2] Ntt Corporation, 1-1 Hikari-no-oka, , Kanagawa, Yokosuka,239-0847, Japan

来源：

2022 European Conference on Optical Communication, ECOC 2022 | 2022年

关键词：

D O I：

暂无

中图分类号：

学科分类号：

摘要：

Capacity enhancement - Congestion-aware - Design and control - Embeddings - First fit - Input parameter - Link utilization - Network design - Network-control - Reinforcement learnings

引用