arXiv subjects

Balázs Varga

Publications and source records attributed to Balázs Varga.

5 recordsLinked to original sources

arXiv2026

Compensating the Packet Delay Variation for 6G Integrated with IEEE Time-Sensitive Networking

6G is deemed as a key technology to support emerging applications with stringent requirements for highly dependable and timecritical communication. In this paper, we investigate 6G networks integrated with TSN and how to compensate for wireless stochastic behavior which involves a large intrinsic packet delay variation. We evaluate a 6G solution to reduce packet delay variation that is based on de-jittering. For this, we propose to use virtual timeslots for providing the required time-awareness. We discuss the benefits of the proposed solution while evaluating the impact of the timeslot size on the number of schedulable TSN streams.

cs.NI↗

arXiv2023

Lightweight Implementation of Per-packet Service Protection in eBPF/XDP

Deterministic communication means reliable packet forwarding with close to zero packet loss and bounded latency. Packet loss or delay above a threshold caused by, e.g., equipment failure or malfunction could be catastrophic for applications that require deterministic communication. To meet loss related targets, per-packet service protection has been introduced by deterministic communications standards; it is provided by Frame Replication and Elimination for Reliability (FRER) for Layer 2 Ethernet networks and by Packet Replication, Elimination, and Ordering Functions (PREOF) for Layer 3 IP/MPLS networks. We have implemented FRER with two conceptually different methods: (1) in eBPF/XDP as a lightweight software implementation; and (2) in userspace. We evaluate our XDP FRER via an experimental analysis and compare the two FRER implementations.

cs.NI↗

arXiv2023

Controlled Descent Training

In this work, a novel and model-based artificial neural network (ANN) training method is developed supported by optimal control theory. The method augments training labels in order to robustly guarantee training loss convergence and improve training convergence rate. Dynamic label augmentation is proposed within the framework of gradient descent training where the convergence of training loss is controlled. First, we capture the training behavior with the help of empirical Neural Tangent Kernels (NTK) and borrow tools from systems and control theory to analyze both the local and global training dynamics (e.g. stability, reachability). Second, we propose to dynamically alter the gradient descent training mechanism via fictitious labels as control inputs and an optimal state feedback policy. In this way, we enforce locally $\mathcal{H}_2$ optimal and convergent training behavior. The novel algorithm, \textit{Controlled Descent Training} (CDT), guarantees local convergence. CDT unleashes new potentials in the analysis, interpretation, and design of ANN architectures. The applicability of the method is demonstrated on standard regression and classification problems.

math.OC↗

arXiv2022

Constrained Policy Gradient Method for Safe and Fast Reinforcement Learning: a Neural Tangent Kernel Based Approach

This paper presents a constrained policy gradient algorithm. We introduce constraints for safe learning with the following steps. First, learning is slowed down (lazy learning) so that the episodic policy change can be computed with the help of the policy gradient theorem and the neural tangent kernel. Then, this enables us the evaluation of the policy at arbitrary states too. In the same spirit, learning can be guided, ensuring safety via augmenting episode batches with states where the desired action probabilities are prescribed. Finally, exogenous discounted sum of future rewards (returns) can be computed at these specific state-action pairs such that the policy network satisfies constraints. Computing the returns is based on solving a system of linear equations (equality constraints) or a constrained quadratic program (inequality constraints, regional constraints). Simulation results suggest that adding constraints (external information) to the learning can improve learning in terms of speed and transparency reasonably if constraints are appropriately selected. The efficiency of the constrained learning was demonstrated with a shallow and wide ReLU network in the Cartpole and Lunar Lander OpenAI gym environments. The main novelty of the paper is giving a practical use of the neural tangent kernel in reinforcement learning.

cs.LG↗

arXiv2019

Multi timescale bandwidth profile and its application for burst-aware fairness

We propose a resource sharing scheme that takes into account the traffic history over several predefined time scales and provides fair resource sharing considering the traffic history. Our concept builds on a simplified version of core-stateless resource sharing, where we only use a few Drop Precedences (DPs). For packet marking we introduce Multi timescale bandwidth profile. Additionally, we provide basic dimensioning concepts for the proposed schema and present its simulation based performance analysis.

cs.NI↗