RL-DE: Reinforcement Learning-Controlled Differential Evolution Algorithm Results

Results from an ablation experiment for the RL-DE algorithm, a reinforcement learning-controlled differential evolution method. The data, shared by Yang Cao on figshare, was last updated on May 13, 2026. The algorithm was evaluated on the CEC2017 benchmark test set and applied to a flexible job shop scheduling problem.

Use Cases

• Benchmarking adaptive differential evolution algorithms based on the CEC2017 test results.
• Analyzing the impact of parameter and strategy selection on algorithm performance based on the described ablation experiment.
• Studying the application of reinforcement learning to control optimization algorithm parameters.
• Investigating the generalization of continuous optimization frameworks to discrete combinatorial problems like job shop scheduling.

Strengths

• Data is associated with a peer-reviewed publication describing a novel three-layer RL-DE framework.
• Algorithm performance was validated on the standard CEC2017 benchmark and a real-world scheduling problem.

Limitations

• The dataset is very small at 5.5 KB, indicating limited scope or a highly summarized result set.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Yang Cao

Collection Method

Results generated from computational experiments with the RL-DE algorithm.

Freshness

Last updated 2026-05-13 17:45:13; freshness should be verified.