描述强化学习

  1. As an important class of machine learning methods, RL aims to solve uncertain decision-making problems by interacting with the environment and near-optimal or suboptimal policies can be obtained in a data-driven way [24].
  2. Reinforcement learning techniques [24] address the problem of how an agent can learn to approximate an optimal or near-optimal behavioral strategy while interacting with its environment.

Jun Wu, Xin Xu, Pengcheng Zhang, Chunming Liu, A novel multi-agent reinforcement learning approach for job scheduling in Grid computing, Future Generation Computer Systems, Volume 27, Issue 5, 2011, Pages 430-439, ISSN 0167-739X, https://doi.org/10.1016/j.future.2010.10.009. (https://www.sciencedirect.com/science/article/pii/S0167739X10002025)

Reinforcement Learning (RL) is a type of machine learning technique that enables an agent to learn in an interactive environment by trial and error using feedback from its own actions and experiences. Different from supervised learning where feedback sent to the agent is a set of correct actions for performing a task, RL uses rewards and punishments as return signals for its positive and negative behavior.

Li, H., Huang, J., Wang, B. et al. Weighted double deep Q-network based reinforcement learning for bi-objective multi-workflow scheduling in the cloud. Cluster Comput 25, 751–768 (2022). https://doi.org/10.1007/s10586-021-03454-6

A reinforcement-learning agent is modeled to perform sequential decision-making by interacting with the environment.

好句

However, the question of how best to exploit the opportunity for centralised learning remains open.

Foerster J, Farquhar G, Afouras T, et al. Counterfactual multi-agent policy gradients[C]//Proceedings of the AAAI conference on artificial intelligence. 2018, 32(1).