Deep LSTM and LSTM-Attention Q-learning based reinforcement learning in oil and gas sector prediction.
David Opeoluwa OyewolaSulaiman Awwal AkinwunmiTemidayo Oluwatosin OmotehinwaPublished in: Knowl. Based Syst. (2024)
Keyphrases
- reinforcement learning
- recurrent neural networks
- function approximation
- reinforcement learning algorithms
- state space
- model free
- natural gas
- temporal difference learning
- prediction accuracy
- learning algorithm
- multi agent
- optimal policy
- multi agent reinforcement learning
- action selection
- markov decision processes
- neural network
- oil field
- eligibility traces
- genetic algorithm
- stochastic approximation
- temporal difference
- control problems
- reinforcement learning methods
- machine learning
- short term prediction
- cooperative
- learning problems
- learning agent
- reward function
- visual attention
- continuous state and action spaces
- relational reinforcement learning
- learning process
- higher education
- continuous state
- partially observable
- feed forward
- learning agents
- state action
- policy iteration
- single agent