Multi-Agent Deep Reinforcement Learning With Progressive Negative Reward for Cryptocurrency Trading.
Kittiwin KumlungmakPeerapon VateekulPublished in: IEEE Access (2023)
Keyphrases
- reinforcement learning
- multi agent
- state space
- reinforcement learning algorithms
- positive and negative
- function approximation
- intelligent agents
- learning algorithm
- cooperative
- markov decision processes
- multiagent systems
- optimal policy
- machine learning
- eligibility traces
- learning process
- electronic commerce
- supervised learning
- single agent
- total reward
- optimal control
- reinforcement learning methods
- learning agents
- temporal difference
- model free
- reward function
- action selection
- reinforcement learning agents
- multi agent environments
- state abstraction
- heterogeneous agents
- partially observable environments
- multi agent systems
- dynamic programming
- autonomous agents
- software agents
- trading strategies
- stock exchange
- cognitive agents
- trading systems
- planning problems
- financial markets