{
  "id": "2108.03952",
  "version": "v1",
  "published": "2021-08-09T11:33:46.000Z",
  "updated": "2021-08-09T11:33:46.000Z",
  "title": "Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces",
  "authors": [
    "Ziyad Sheebaelhamd",
    "Konstantinos Zisis",
    "Athina Nisioti",
    "Dimitris Gkouletsos",
    "Dario Pavllo",
    "Jonas Kohler"
  ],
  "comment": "ICML 2021 Workshop on Reinforcement Learning for Real Life",
  "categories": [
    "cs.LG",
    "cs.RO"
  ],
  "abstract": "Multi-agent control problems constitute an interesting area of application for deep reinforcement learning models with continuous action spaces. Such real-world applications, however, typically come with critical safety constraints that must not be violated. In order to ensure safety, we enhance the well-known multi-agent deep deterministic policy gradient (MADDPG) framework by adding a safety layer to the deep policy network. %which automatically corrects invalid actions. In particular, we extend the idea of linearizing the single-step transition dynamics, as was done for single-agent systems in Safe DDPG (Dalal et al., 2018), to multi-agent settings. We additionally propose to circumvent infeasibility problems in the action correction step using soft constraints (Kerrigan & Maciejowski, 2000). Results from the theory of exact penalty functions can be used to guarantee constraint satisfaction of the soft constraints under mild assumptions. We empirically find that the soft formulation achieves a dramatic decrease in constraint violations, making safety available even during the learning procedure.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2021-08-09T11:33:46.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "safe deep reinforcement learning",
      "continuous action spaces",
      "multi-agent systems",
      "multi-agent deep deterministic policy",
      "deep deterministic policy gradient"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}