Jul 2026· International journal of information and communication technology trends· Vol 6, pp. 28-41· 0 citations
TL;DR
A reproducible discrete-time link-state convergence proxy with independent topology views, stochastic failure detection and link-state advertisement diffusion, shortest-path recomputation, and forwarding-loop/drop checks is implemented.
Abstract
Routing failures demand rapid control-plane convergence, yet uniformly aggressive detection and flooding increases signaling load and can induce unstable parameter oscillations. This paper presents MARL-Converge, a decentralized multi-agent reinforcement learning framework in which every router selects a local conservative, balanced, or aggressive control profile from router-local and one-hop telemetry. A parameter-shared dueling Double Deep Q-Network is trained with shared experience and a global-dominant team reward, while execution remains decentralized. A failure-aware action mask prevents routers without local evidence from blindly escalating to the aggressive profile. We implement a reproducible discrete-time link-state convergence proxy with independent topology views, stochastic failure detection and link-state advertisement diffusion, shortest-path recomputation, and forwarding-loop/drop checks. Evaluation comprises 600 matched episodes on the real Abilene and held-out AARNet topologies from the Internet Topology Zoo. Relative to a static balanced profile, MARL-Converge reduces convergence steps by 19.0% and modeled packet loss by 83.6%; the 6.1% overhead reduction is directionally favorable but not significant at the 0.05 level. Relative to a local heuristic, it is 9.3% faster, uses 14.0% fewer control units, and makes 56.4% fewer profile switches. It is 7.6% slower than always-aggressive control but cuts overhead by 35.4%. These results establish a measured speed-stability-efficiency trade-off without claiming physical-router timing equivalence.
Timely delivery of delay-sensitive information over dynamic, heterogeneous networks is essential for NextG interactive applications, yet providing strict End-to-End (E2E) peak latency guarantees remains an open challenge. Two obstacles limit the adoption of learning-based network control in this setting: traditional volume-based routing metrics, while highly effective for general traffic management, are not designed to capture traffic urgency; and Deep Reinforcement Learning (DRL) controllers trained from scratch suffer from sample inefficiency, long training times, and early-stage exploration volatility. This paper introduces a deployment-focused network control framework that addresses both obstacles. First, we present Effective Congestion (EC), a deadline-aware metric family that quantifies interface congestion by packet urgency and proactively filters non-viable traffic, coupled with a Uniform Path Grouping (UPG) distribution heuristic promoting robust load-balancing; the resulting policies are embedded into Multi-Agent Deep Reinforcement Learning Effective Congestion ($p^*$) (MADRL EC ($p^*$)), a hybrid architecture combining a distributed scheduler with a centralized RL-based router. Second, we introduce a unified training objective that generalizes existing policy-learning paradigms---behavioral cloning, offline Reinforcement Learning (RL), online RL, and offline-to-online schemes---as special cases, combining a live-reward term, a pre-collected-reward term, and a policy-imitation term. From this objective, we derive the Model-Guided Annealed Reinforcement Learning (MGA-RL) protocol, instantiated on a Deep Deterministic Policy Gradient (DDPG) backbone: a deployment-oriented, demonstration-driven training approach that generalizes conventional Offline-to-Online (O2O) schemes, in which trajectories from a lightweight [...]
Vincenzo Norman Vitale, Mohammad Solki, A. Tulino et al.· 0 citations
The results support conformal shielding as a practical mechanism for exposing and controlling the safety-speed trade-off in learning-based routing, while also identifying the limits of guarantees under topology and load shift.
Fatima Rahman, Eric Nolan· Academic Journal of Applied...· 0 citations
GraphRoute-Transfer, a graph-neural-network policy that assigns per-node timers from local structural features and is by construction permutation- and size-invariant, is proposed, a graph-neural-network policy that assigns per-node timers from local structural features and is by construction permutation- and size-invariant.
Yuto Nakamura· Journal of Computing and Ele...· 0 citations
GraphRoute-Transfer, a graph-neural-network policy that assigns per-node timers from local structural features and is by construction permutation-and size-invariant, is proposed, a graph-neural-network policy that assigns per-node timers from local structural features and is by construction permutation-and size-invariant.
Future sixth-generation (6G)-oriented networks require programmable control that can adapt routing to latency and congestion without unsafe online exploration. This study evaluates offline multi-agent deep deterministic policy gradient (MADDPG) with behavior-adjusted training rewards for latency-aware path control in software-defined networking (SDN). Each traffic pair is modeled as an agent selecting one of three retained candidate paths, while centralized critics learn coordinated decisions from topology-specific Ryu–Mininet transition datasets. Nine policies are compared using ten paired seeds on fat-tree, mesh-grid, and WAN-corridors topologies under a deployed utilization–latency weighting of 0.60/0.40, together with flow-completion, latency, congestion, architectural-comparison, sensitivity, robustness, statistical, and controller-overhead analyses. The utilization-aware path heuristic achieves the strongest overall reward ranking. MADDPG is the strongest learned policy on fat-tree, is not significantly outperformed by any evaluated policy on mesh-grid, and remains statistically tied with completion-matched policies on WAN-corridors. Behavior adjustment is topology-dependent rather than uniformly beneficial. The exported policy requires approximately 52μs per joint decision, whereas complete control-loop timing is dominated by network-statistics polling. These results support offline multi-agent SDN control as a competitive, low-overhead option when interpreted jointly with topology structure, flow completion, and strong heuristic baselines.
A. Kyzyrkanov, Y. Nurakhov, Zhenis Otarbay et al.· Technologies· 0 citations
Delay Tolerant Networks (DTNs) operate under intermittent connectivity, dynamic topology, and resource-constrained nodes, making efficient routing a persistent challenge. Many existing DTN routing protocols—including probabilistic, social-aware, and utility-based approaches—rely on static or semi-static control parameters that do not adjust to changing network conditions. In our earlier work, PASTOR-DTN introduced a five-signal utility framework integrating encounter predictability, social trust, centrality, buffer headroom, and TTL urgency with token-bucket–based forwarding control. However, the utility weights and token refill rate in PASTOR-DTN remain fixed, limiting responsiveness to time-varying network dynamics. This paper presents PASTOR-Adaptive, a bounded feedback-control framework that extends PASTOR-DTN through joint online adaptation of utility weights and forwarding rate. Rather than employing learning-based optimization with large state spaces and convergence delays, the proposed framework applies a lightweight deterministic control law that adjusts parameters according to locally observable network phase (sparse, normal, or dense) and congestion state. Stability properties are maintained through bounded update increments, normalized weight projection, and negative feedback regulation. An ablation study on five different network settings (sparse, normal, dense, high mobility, heavy traffic) reveals that token rate adaptation is the main adaptive component, reducing the overhead by 29% for dense settings and increasing forwarding activity by 42% for sparse settings. Sensitivity analysis demonstrates robustness, with delivery probability varying by less than 0.5% across a tenfold parameter range. Comparative evaluation against Epidemic, PRoPHET, Spray-and-Wait, MaxProp, and PASTOR-DTN over 18 scenarios indicates that PASTOR-Adaptive achieves 84–98% delivery with overhead ratios between 1.3 and 4.1. These results demonstrate that bounded deterministic adaptation can maintain stable delivery–overhead trade-offs across heterogeneous DTN conditions.
Lakshmi Narayana· i-manager's Journal on Compu...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
