@unpublished{30790, abstract = {{Iterative distributed optimization algorithms involve multiple agents that communicate with each other, over time, in order to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.}}, author = {{Redder, Adrian and Ramaswamy, Arunselvan and Karl, Holger}}, booktitle = {{arXiv:2201.11343}}, title = {{{Distributed gradient-based optimization in the presence of dependent aperiodic communication}}}, year = {{2022}}, } @unpublished{30791, abstract = {{We present sufficient conditions that ensure convergence of the multi-agent Deep Deterministic Policy Gradient (DDPG) algorithm.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.}}, author = {{Redder, Adrian and Ramaswamy, Arunselvan and Karl, Holger}}, booktitle = {{arXiv:2201.00570}}, title = {{{Asymptotic Convergence of Deep Multi-Agent Actor-Critic Algorithms}}}, year = {{2022}}, } @techreport{33854, abstract = {{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.}}, author = {{Schneider, Stefan Balthasar and Karl, Holger and Khalili, Ramin and Hecker, Artur}}, keywords = {{mobility management, coordinated multipoint, CoMP, cell selection, resource management, reinforcement learning, multi agent, MARL, self-learning, self-adaptation, QoE}}, title = {{{DeepCoMP: Coordinated Multipoint Using Multi-Agent Deep Reinforcement Learning}}}, year = {{2021}}, } @techreport{35889, abstract = {{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.}}, author = {{Schneider, Stefan Balthasar}}, keywords = {{nfv, coordination, machine learning, reinforcement learning, phd, digest}}, title = {{{Conventional and Machine Learning Approaches for Network and Service Coordination}}}, year = {{2021}}, } @techreport{2483, abstract = {{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.}}, author = {{Dräxler, Sevil and Peuster, Manuel and Illian, Marvin and Karl, Holger}}, title = {{{Towards Predicting Resource Demands and Performance of Distributed Cloud Services}}}, year = {{2018}}, } @techreport{6485, author = {{Rosa, Raphael Vicente and Rothenberg, Christian Esteve and Peuster, Manuel and Karl, Holger}}, publisher = {{IETF}}, title = {{{Methodology for VNF Benchmarking Automation}}}, year = {{2018}}, } @unpublished{749, author = {{Dräxler, Sevil and Karl, Holger}}, booktitle = {{CoRR}}, title = {{{Specification of Complex Structures in Distributed Service Function Chaining Using a YANG Data Model}}}, year = {{2015}}, } @unpublished{750, author = {{Dräxler, Martin and Karl, Holger}}, booktitle = {{CoRR}}, title = {{{Dynamic Backhaul Network Configuration in SDN-based Cloud RANs}}}, year = {{2015}}, } @unpublished{766, author = {{Mehraghdam, Sevil and Keller, Matthias and Karl, Holger}}, booktitle = {{CoRR}}, title = {{{Specifying and Placing Chains of Virtual Network Functions}}}, year = {{2014}}, } @unpublished{767, author = {{Wette, Philip and Karl, Holger}}, booktitle = {{CoRR}}, title = {{{DCT²Gen: A Versatile TCP Traffic Generator for Data Centers}}}, year = {{2014}}, }