Sie haben Javascript deaktiviert!
Sie haben versucht eine Funktion zu nutzen, die nur mit Javascript möglich ist. Um sämtliche Funktionalitäten unserer Internetseite zu nutzen, aktivieren Sie bitte Javascript in Ihrem Browser.

Diese Seite ist nicht in Deutsch verfügbar

[Translate to Deutsch:] Foto: Judith Kraft

Liste im Research Information System öffnen

Distributed gradient-based optimization in the presence of dependent aperiodic communication

A. Redder, A. Ramaswamy, H. Karl, in: arXiv:2201.11343, 2022

Iterative distributed optimization algorithms involve multiple agents that communicate with each other, over time, in order to minimize/maximize a global objective. In the presence of unreliable communication networks, the Age-of-Information (AoI), which measures the freshness of data received, may be large and hence hinder algorithmic convergence. In this paper, we study the convergence of general distributed gradient-based optimization algorithms in the presence of communication that neither happens periodically nor at stochastically independent points in time. We show that convergence is guaranteed provided the random variables associated with the AoI processes are stochastically dominated by a random variable with finite first moment. This improves on previous requirements of boundedness of more than the first moment. We then introduce stochastically strongly connected (SSC) networks, a new stochastic form of strong connectedness for time-varying networks. We show: If for any $p \ge0$ the processes that describe the success of communication between agents in a SSC network are $\alpha$-mixing with $n^{p-1}\alpha(n)$ summable, then the associated AoI processes are stochastically dominated by a random variable with finite $p$-th moment. In combination with our first contribution, this implies that distributed stochastic gradient descend converges in the presence of AoI, if $\alpha(n)$ is summable.

Asymptotic Convergence of Deep Multi-Agent Actor-Critic Algorithms

A. Redder, A. Ramaswamy, H. Karl, in: arXiv:2201.00570, 2022

We present sufficient conditions that ensure convergence of the multi-agent Deep Deterministic Policy Gradient (DDPG) algorithm. It is an example of one of the most popular paradigms of Deep Reinforcement Learning (DeepRL) for tackling continuous action spaces: the actor-critic paradigm. In the setting considered herein, each agent observes a part of the global state space in order to take local actions, for which it receives local rewards. For every agent, DDPG trains a local actor (policy) and a local critic (Q-function). The analysis shows that multi-agent DDPG using neural networks to approximate the local policies and critics converge to limits with the following properties: The critic limits minimize the average squared Bellman loss; the actor limits parameterize a policy that maximizes the local critic's approximation of $Q_i^*$, where $i$ is the agent index. The averaging is with respect to a probability distribution over the global state-action space. It captures the asymptotics of all local training processes. Finally, we extend the analysis to a fully decentralized setting where agents communicate over a wireless network prone to delays and losses; a typical scenario in, e.g., robotic applications.

DeepCoMP: Coordinated Multipoint Using Multi-Agent Deep Reinforcement Learning

S.B. Schneider, H. Karl, R. Khalili, A. Hecker, 2021

Macrodiversity is a key technique to increase the capacity of mobile networks. It can be realized using coordinated multipoint (CoMP), simultaneously connecting users to multiple overlapping cells. Selecting which users to serve by how many and which cells is NP-hard but needs to happen continuously in real time as users move and channel state changes. Existing approaches often require strict assumptions about or perfect knowledge of the underlying radio system, its resource allocation scheme, or user movements, none of which is readily available in practice. Instead, we propose three novel self-learning and self-adapting approaches using model-free deep reinforcement learning (DRL): DeepCoMP, DD-CoMP, and D3-CoMP. DeepCoMP leverages central observations and control of all users to select cells almost optimally. DD-CoMP and D3-CoMP use multi-agent DRL, which allows distributed, robust, and highly scalable coordination. All three approaches learn from experience and self-adapt to varying scenarios, reaching 2x higher Quality of Experience than other approaches. They have very few built-in assumptions and do not need prior system knowledge, making them more robust to change and better applicable in practice than existing approaches.

Conventional and Machine Learning Approaches for Network and Service Coordination

S.B. Schneider, 2021

Network and service coordination is important to provide modern services consisting of multiple interconnected components, e.g., in 5G, network function virtualization (NFV), or cloud and edge computing. In this paper, I outline my dissertation research, which proposes six approaches to automate such network and service coordination. All approaches dynamically react to the current demand and optimize coordination for high service quality and low costs. The approaches range from centralized to distributed methods and from conventional heuristic algorithms and mixed-integer linear programs to machine learning approaches using supervised and reinforcement learning. I briefly discuss their main ideas and advantages over other state-of-the-art approaches and compare strengths and weaknesses.

Towards Predicting Resource Demands and Performance of Distributed Cloud Services

S. Dräxler, M. Peuster, M. Illian, H. Karl, 2018

Understanding the behavior of distributed cloud service components in different load situations is important for efficient and automatic management and orchestration of these services. For this purpose and for practical research in distributed cloud computing in general, there is need for benchmarks and experimental data. In this paper, we describe our experiments for characterizing the relationship between resource demands of application components and the expected performance of applica- tions. We present initial results for predicting the interdependence between resource demands and performance characteristics using support vector regression and polynomial regression models. The data gathered from our experiments is publicly available.

Methodology for VNF Benchmarking Automation

R.V. Rosa, C.E. Rothenberg, M. Peuster, H. Karl, IETF, 2018

Specifying and Placing Chains of Virtual Network Functions

S. Mehraghdam, M. Keller, H. Karl, in: CoRR, 2014

Adding Geographical Embedding to AS Topology Generation

A. Schwabe, H. Karl, in: CoRR, 2014

A Scalable Redundant TDMA Protocol for High-Density WSNs Inside an Aircraft

J. Blanckenstein, J. Garcia-Jimenez, J. Klaue, H. Karl, in: Lecture Notes in Electrical Engineering, Springer International Publishing, 2013, pp. 165-177

Denser Networks for the Future Internet, the CROWD Approach

A. de la Oliva, A. Morelli, V. Mancuso, M. Draexler, T. Hentschel, T. Melia, P. Seite, C. Cicconetti, in: Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, Springer Berlin Heidelberg, 2013, pp. 28-41

Anticipatory Buffer Control and Quality Selection for Wireless Video Streaming

M. Dräxler, J. Blobel, P. Dreimann, S. Valentin, H. Karl, in: CoRR, 2013

Introducing feedback to preemptive routing and wavelength assignment algorithms for dynamic traffic scenarios

P. Wette, H. Karl, Universität Paderborn, 2012

Preemptive Routing and Wavelength Assignment (RWA) algorithms preempt established lightpaths in case not enough resources are available to setup a new lightpath in a Wavelength Division Multiplexing (WDM) network. The selection of lightpaths to be preempted relies on internal decisions of the RWA algorithm. Thus, if dedicated properties of the network topology are required by the applications running on the network, these requirements have to be known by the RWA algorithm. Otherwise it might happen that by preempting a particular lightpath these requirements are violated. If, however, these requirements include parameters only known at the nodes running the application, the RWA algorithm cannot evaluate the requirements. For this reason a RWA algorithm is needed which involves its users in the preemption decisions. We present a family of preemptive RWA algorithms for WDM networks. These algorithms have two distinguishing features: a) they can handle dynamic trafﬁc by on-the-ﬂy reconﬁguration, and b) users can give feedback for reconﬁguration decisions and thus inﬂuence the preemption decision of the RWA algorithm, leading to networks which adapt directly to application needs. This is different from trafﬁc engineering where the network is (slowly) adapted to observed trafﬁc patterns. Our algorithms handle various WDM network conﬁgurations including networks consisting of heterogeneous WDM hardware. To this end, we are using the layered graph approach together with a newly developed graph model that is used to determine conﬂicting lightpaths.

How to manage and Search/Retrieve Information Objects

C. Dannewitz, e. al, in: Architecture and Design for the Future Internet, 2011

Integrating Generic Paths and NetInf

C. Dannewitz, e. al, in: Architecture and Design for the Future Internet, 2011

Scenarios, Research Issues, and Architecture for Ubiquitous Sensing

T. Kanter, V. Kardeby, S. Forsström, J. Walters, in: Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, Springer Berlin Heidelberg, 2011, pp. 285-297

Selected properties of a joint congestion controller for TCP connections

M. Savorić, H. Karl, A. Wolisz, in: Providing Quality of Service in Heterogeneous Environments, Proceedings of the 18th International Teletraffic Congress - ITC-18, Elsevier, 2010, pp. 861-870

Creating Butterflies in the Core – A Network Coding Extension for MPLS/RSVP-TE

T. Biermann, A. Schwabe, H. Karl, in: NETWORKING 2009, Springer Berlin Heidelberg, 2009, pp. 883-894

Network-Coding-Based Cooperative Transmission in Wireless Sensor Networks: Diversity-Multiplexing Tradeoff and Coverage Area Extension

D.H. Woldegebreal, H. Karl, in: Lecture Notes in Computer Science, Springer Berlin Heidelberg, 2008, pp. 141-155

Experimental evaluation of IEEE 802.11a-based WLANs for medium range communication

F. Eitzen, S. Valentin, K. Gossens, H. Karl, O. Rolfes, 2007

H. Karl, in: The Industrial Information Technology Handbook, 2005, pp. 1--16

Data Transmission in Mobile Communication Systems

H. Karl, in: Location-Based Services, 2004, pp. 207--244

A common wireless sensor network architecture?

V. Handziski, A. Köpke, H. Karl, A. Wolisz, 2003

Relaying in Wireless Access Networks

H. Karl, S. Mengesha, D. Hollos, Business Briefing: Wireless Technology 2002, 2002

Liste im Research Information System öffnen

##### Sie interessieren sich für:

Die Universität der Informationsgesellschaft