AF-TT-RDQN: Network Parameters for Maritime Edge Computing Offloading

Network parameters for the Action-Factorized Rainbow DQN with Token Transformer algorithm, authored by Shengtian Zhang and uploaded to figshare in May 2026. The 5.5 KB dataset contains experimental results for a deep reinforcement learning method designed to optimize computation offloading strategies in Edge Computing-enabled Internet of Ships environments.

Use Cases

• Benchmarking new reinforcement learning algorithms for edge computing based on the reported latency-energy performance.
• Analyzing the effectiveness of action factorization mechanisms in handling high-dimensional decision spaces as described.
• Studying strategies to achieve zero invalid action rates in constrained maritime wireless systems.
• Reproducing the experimental results for the weighted latency-energy objective optimization.

Strengths

• Dataset is openly licensed under CC-BY-4.0.
• The associated study reports the algorithm achieves a zero rate of invalid actions and satisfies all system constraints.
• File size is precisely reported as 5.5 KB.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Shengtian Zhang via figshare.

Collection Method

Likely generated from experiments evaluating the proposed AF-TT-RDQN algorithm.

Freshness

Last updated 2026-05-11 17:37:59; freshness should be verified.