Using a multi-agent transformer where each TSN queue is an agent cuts simulated XR deadline misses and queueing latency compared to PPO, A2C, HAPPO, and backlog or deadline heuristics.
Self-play ensemble q-learning enabled resource allocation for network slicing,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multi-Agent Transformer for Queue-Level XR Traffic Scheduling in TSN Networks
Using a multi-agent transformer where each TSN queue is an agent cuts simulated XR deadline misses and queueing latency compared to PPO, A2C, HAPPO, and backlog or deadline heuristics.