Dynamic multi-objective optimisation using deep reinforcement learning: benchmark, algorithm and an application to identify vulnerable zones based on water quality

Keyphrases

• Dynamic Environment 100%
• Reinforcement Learning 100%
• Zone-based 100%
• Water Quality 100%
• Deep Reinforcement Learning (deep RL) 100%
• Benchmark Algorithms 100%
• Vulnerable Zones 100%
• Dynamic multi-objective Optimization 100%
• Learning Research 66%
• Dynamic multi-objective Problems 66%
• Deep Q Network Algorithm 66%
• Brazil 33%
• Pareto Front 33%
• Multi-objective Optimization Problem 33%
• Multi-objective Optimization 33%
• Learning Settings 33%
• Resilience 33%
• Expert System 33%
• Constrained Environment 33%
• Dynamic Nature 33%
• Function Problems 33%
• Function Constraints 33%
• Deep Sea 33%
• Objective Constraints 33%
• Generic Algorithm 33%
• Compromise Solution 33%
• Markov Decision Process 33%
• Treasure Hunt 33%

Computer Science

• Multi-Objective Optimization 100%
• Testbed 100%
• Deep Reinforcement Learning 100%
• Reinforcement Learning 75%
• Dynamic Environment 75%
• Multiobjective 50%
• Optimization Problem 50%
• Network Algorithm 50%
• Deep Q-Network 50%
• Objective Function 25%
• Markov Decision Process 25%
• Modified Version 25%
• Dynamic Nature 25%
• Expert System 25%
• Generic Algorithm 25%
• Constrained Environment 25%

Engineering

• Markov Decision Process 20%