Deep Reinforcement Learning with Successive Over-Relaxation and its Application in Autoscaling Cloud Resources

John, Indu ; Bhatnagar, Shalabh (2020) Deep Reinforcement Learning with Successive Over-Relaxation and its Application in Autoscaling Cloud Resources In: International Joint Conference on Neural Networks (IJCNN), 19-24 July 2020, Glasgow, UK.

Full text not available from this repository.

Official URL: http://doi.org/10.1109/IJCNN48605.2020.9206598

Related URL: http://dx.doi.org/10.1109/IJCNN48605.2020.9206598

Abstract

We present a new deep reinforcement learning algorithm using the technique of successive over-relaxation (SOR) in Deep Q-networks (DQNs). The new algorithm, named SOR-DQN, uses modified targets in the DQN framework with the aim of accelerating training. This work is motivated by the problem of auto-scaling resources for cloud applications, for which existing algorithms suffer from issues such as slow convergence, poor performance during the training phase and non-scalability. For the above problem, SOR-DQN achieves significant improvements over DQN on both synthetic and real datasets. We also study the generalization ability of the algorithm to multiple tasks by using it to train agents playing Atari video games.

Item Type:Conference or Workshop Item (Paper)
Source:Copyright of this article belongs to Institute of Electrical and Electronics Engineers.
Keywords:Reinforcement Learning; Deep Learning; Cloud Computing; Resource Allocation; Atari Games.
ID Code:116614
Deposited On:12 Apr 2021 07:09
Last Modified:12 Apr 2021 07:09

Repository Staff Only: item control page