Open access
2026
Decentralized Task Scheduling in Distributed Systems: A Lightweight Multi-Agent Deep Reinforcement Learning Approach With Gossip-Based Consensus
DRL-MADRL is competitive with the strongest heuristic under low contention and achieves the best SLA satisfaction at moderate and high contention, and the NumPy implementation requires approximately 80 KB per agent and sub-10 ms inference latency.
Daniel Benniah John
· IEEE Access · 0 citations