arXiv subjects
Publications and source records attributed to George Iosifidis.
We show that a kernel estimator using multiple function evaluations can be easily converted into a sampling-based bandit estimator with expectation equal to the original kernel estimate. Plugging such a bandit estimator into the standard FTRL algorithm yields a bandit convex optimisation algorithm that achieves $\tilde{O}(t^{1/2})$ regret against adversarial time-varying convex loss functions.
The virtualization of wireless networks enables new services to access network resources made available by the Network Operator (NO) through a Network Slicing market. The different service providers (SPs) have the opportunity to lease the network resources from the NO to constitute slices that address the demand of their specific network service. The goal of any SP is to maximize its service utility and minimize costs from leasing resources while facing uncertainties of the prices of the resources and the users' demand. In this paper, we propose a solution that allows the SP to decide its online reservation policy, which aims to maximize its service utility and minimize its cost of reservation simultaneously. We design the Optimistic Online Learning for Reservation (OOLR) solution, a decision algorithm built upon the Follow-the-Regularized Leader (FTRL), that incorporates key predictions to assist the decision-making process. Our solution achieves a $\mathcal{O}(\sqrt{T})$ regret bound where $T$ represents the horizon. We integrate a prediction model into the OOLR solution and we demonstrate through numerical results the efficacy of the combined models' solution against the FTRL baseline.
We propose an ML-based model that automates and expedites the solution of MIPs by predicting the values of variables. Our approach is motivated by the observation that many problem instances share salient features and solution structures since they differ only in few (time-varying) parameters. Examples include transportation and routing problems where decisions need to be re-optimized whenever commodity volumes or link costs change. Our method is the first to exploit the sequential nature of the instances being solved periodically, and can be trained with ``unlabeled'' instances, when exact solutions are unavailable, in a semi-supervised setting. Also, we provide a principled way of transforming the probabilistic predictions into integral solutions. Using a battery of experiments with representative binary MIPs, we show the gains of our model over other ML-based optimization approaches.
Recommendations are employed by Content Providers (CPs) of streaming services in order to boost user engagement and their revenues. Recent works suggest that nudging recommendations towards cached items can reduce operational costs in the caching networks, e.g., Content Delivery Networks (CDNs) or edge cache providers in future wireless networks. However, cache-friendly recommendations could deviate from users' tastes, and potentially affect the CP's revenues. Motivated by real-world business models, this work identifies the misalignment of the financial goals of the CP and the caching network provider, and presents a network-economic framework for recommendations. We propose a cooperation mechanism leveraging the Nash bargaining solution that allows the two entities to jointly design the recommendation policy. We consider different problem instances that vary on the extent these entities are willing to share their cost and revenue models, and propose two cooperative policies, CCR and DCR, that allow them to make decisions in a centralized or distributed way. In both cases, our solution guarantees reaching a fair and Pareto optimal allocation of the cooperation gains. Moreover, we discuss the extension of our framework towards caching decisions. A wealth of numerical experiments in realistic scenarios show the policies lead to significant gains for both entities.
The design of effective online caching policies is an increasingly important problem for content distribution networks, online social networks and edge computing services, among other areas. This paper proposes a new algorithmic toolbox for tackling this problem through the lens of \emph{optimistic} online learning. We build upon the Follow-the-Regularized-Leader (FTRL) framework, which is developed further here to include predictions for the file requests, and we design online caching algorithms for bipartite networks with pre-reserved or dynamic storage subject to time-average budget constraints. The predictions are provided by a content recommendation system that influences the users viewing activity and hence can naturally reduce the caching network's uncertainty about future requests. We also extend the framework to learn and utilize the best request predictor in cases where many are available. We prove that the proposed {optimistic} learning caching policies can achieve \emph{sub-zero} performance loss (regret) for perfect predictions, and maintain the sub-linear regret bound $O(\sqrt T)$, which is the best achievable bound for policies that do not use predictions, even for arbitrary-bad predictions. The performance of the proposed algorithms is evaluated with detailed trace-driven numerical tests.
The design of Open Radio Access Network (O-RAN) compliant systems for configuring the virtualized Base Stations (vBSs) is of paramount importance for network operators. This task is challenging since optimizing the vBS scheduling procedure requires knowledge of parameters, which are erratic and demanding to obtain in advance. In this paper, we propose an online learning algorithm for balancing the performance and energy consumption of a vBS. This algorithm provides performance guarantees under unforeseeable conditions, such as non-stationary traffic and network state, and is oblivious to the vBS operation profile. We study the problem in its most general form and we prove that the proposed technique achieves sub-linear regret (i.e., zero average optimality gap) even in a fast-changing environment. By using real-world data and various trace-driven evaluations, our findings indicate savings of up to 74.3% in the power consumption of a vBS in comparison with state-of-the-art benchmarks.
One of the key benefits of virtualized radio access networks (vRANs) is network management flexibility. However, this versatility raises previously-unseen network management challenges. In this paper, a learning-based zero-touch vRAN orchestration framework (LOFV) is proposed to jointly select the functional splits and allocate the virtualized resources to minimize the long-term management cost. First, testbed measurements of the behaviour between the users' demand and the virtualized resource utilization are collected using a centralized RAN system. The collected data reveals that there are non-linear and non-monotonic relationships between demand and resource utilization. Then, a comprehensive cost model is proposed that takes resource overprovisioning, declined demand, instantiation and reconfiguration into account. Moreover, the proposed cost model also captures different routing and computing costs for each split. Motivated by our measurement insights and cost model, LOFV is developed using a model-free reinforcement learning paradigm. The proposed solution is constructed from a combination of deep Q-learning and a regression-based neural network that maps the network state and users' demand into split and resource control decisions. Our numerical evaluations show that LOFV can offer cost savings by up to 69\% of the optimal static policy and 45\% of the optimal fully dynamic policy.
The design of effective online caching policies is an increasingly important problem for content distribution networks, online social networks and edge computing services, among other areas. This paper proposes a new algorithmic toolbox for tackling this problem through the lens of optimistic online learning. We build upon the Follow-the-Regularized-Leader (FTRL) framework which is developed further here to include predictions for the file requests, and we design online caching algorithms for bipartite networks with fixed-size caches or elastic leased caches subject to time-average budget constraints. The predictions are provided by a content recommendation system that influences the users viewing activity, and hence can naturally reduce the caching network's uncertainty about future requests. We prove that the proposed optimistic learning caching policies can achieve sub-zero performance loss (regret) for perfect predictions, and maintain the best achievable regret bound $O(\sqrt T)$ even for arbitrary-bad predictions. The performance of the proposed algorithms is evaluated with detailed trace-driven numerical tests.
In this paper we extend the classical Follow-The-Regularized-Leader (FTRL) algorithm to encompass time-varying constraints, through adaptive penalization. We establish sufficient conditions for the proposed Penalized FTRL algorithm to achieve $O(\sqrt{t})$ regret and violation with respect to strong benchmark $\hat{X}^{max}_t$. Lacking prior knowledge of the constraints, this is probably the largest benchmark set that we can reasonably hope for. Our sufficient conditions are necessary in the sense that when they are violated there exist examples where $O(\sqrt{t})$ regret and violation is not achieved. Compared to the best existing primal-dual algorithms, Penalized FTRL substantially extends the class of problems for which $O(\sqrt{t})$ regret and violation performance is achievable.
An increasing number of mobile applications rely on Machine Learning (ML) routines for analyzing data. Executing such tasks at the user devices saves the energy spent on transmitting and processing large data volumes at distant cloud-deployed servers. However, due to memory and computing limitations, the devices often cannot support the required resource-intensive routines and fail to accurately execute the tasks. In this work, we address the problem of edge-assisted analytics in resource-constrained systems by proposing and evaluating a rigorous selective offloading framework. The devices execute their tasks locally and outsource them to cloudlet servers only when they predict a significant performance improvement. We consider the practical scenario where the offloading gain and resource costs are time-varying; and propose an online optimization algorithm that maximizes the service performance without requiring to know this information. Our approach relies on an approximate dual subgradient method combined with a primal-averaging scheme, and works under minimal assumptions about the system stochasticity. We fully implement the proposed algorithm in a wireless testbed and evaluate its performance using a state-of-the-art image recognition application, finding significant performance gains and cost savings.
We consider the general problem of online convex optimization with time-varying additive constraints in the presence of predictions for the next cost and constraint functions. A novel primal-dual algorithm is designed by combining a Follow-The-Regularized-Leader iteration with prediction-adaptive dynamic steps. The algorithm achieves $\mathcal O(T^{\frac{3-β}{4}})$ regret and $\mathcal O(T^{\frac{1+β}{2}})$ constraint violation bounds that are tunable via parameter $β\!\in\![1/2,1)$ and have constant factors that shrink with the predictions quality, achieving eventually $\mathcal O(1)$ regret for perfect predictions. Our work extends the FTRL framework for this constrained OCO setting and outperforms the respective state-of-the-art greedy-based solutions, without imposing conditions on the quality of predictions, the cost functions or the geometry of constraints, beyond convexity.
Millions of sensors, cameras, meters, and other edge devices are deployed in networks to collect and analyse data. In many cases, such devices are powered only by Energy Harvesting(EH) and have limited energy available to analyse acquired data. When edge infrastructure is available, a device has a choice: to perform analysis locally or offload the task to other resource-rich devices such as cloudlet servers. However, such a choice carries a price in terms of consumed energy and accuracy. On the one hand, transmitting raw data can result in a higher energy cost in comparison to the required energy to process data locally. On the other hand, performing data analytics on servers can improve the task's accuracy. Additionally, due to the correlation between information sent by multiple devices, accuracy might not be affected if some edge devices decide to neither process nor send data and preserve energy instead. For such a scenario, we propose a Deep Reinforcement Learning (DRL) based solution capable of learning and adapting the policy to the time-varying energy arrival due to EH patterns. We leverage two datasets, one to model energy an EH device can collect and the other to model the correlation between cameras. Furthermore, we compare the proposed solution performance to three baseline policies. Our results show that we can increase accuracy by 15% in comparison to conventional approaches while preventing outages.
Data centers are increasingly using high-speed circuit switches to cope with the growing demand and reduce operational costs. One of the fundamental tasks of circuit switches is to compute a sparse collection of switching configurations to support a traffic demand matrix. Such a problem has been addressed in the literature with variations of the approach proposed by Birkhoff in 1946 to decompose a doubly stochastic matrix exactly. However, the existing methods are heuristic and do not have theoretical guarantees on how well a collection of switching configurations (i.e., permutations) can approximate a traffic matrix (i.e., a scaled doubly stochastic matrix). In this paper, we revisit Birkhoff's approach and make three contributions. First, we establish the first theoretical bound on the sparsity of Birkhoff's algorithm (i.e., the number of switching configurations necessary to approximate a traffic matrix). In particular, we show that by using a subset of the admissible permutation matrices, Birkhoff's algorithm obtains an $ε$-approximate decomposition with at most $O( \log(1 / ε))$ permutations. Second, we propose a new algorithm, Birkhoff+, which combines the wealth of Frank-Wolfe with Birkhoff's approach to obtain sparse decompositions in a fast manner. And third, we evaluate the performance of the proposed algorithm numerically and study how this affects the performance of a circuit switch. Our results show that Birkhoff+ is superior to previous algorithms in terms of throughput, running time, and number of switching configurations.
Wireless networks used for Internet of Things (IoT) are expected to largely involve cloud-based computing and processing. Softwarised and centralised signal processing and network switching in the cloud enables flexible network control and management. In a cloud environment, dynamic computational resource allocation is essential to save energy while maintaining the performance of the processes. The stochastic features of the Central Processing Unit (CPU) load variation as well as the possible complex parallelisation situations of the cloud processes makes the dynamic resource allocation an interesting research challenge. This paper models this dynamic computational resource allocation problem into a Markov Decision Process (MDP) and designs a model-based reinforcement-learning agent to optimise the dynamic resource allocation of the CPU usage. Value iteration method is used for the reinforcement-learning agent to pick up the optimal policy during the MDP. To evaluate our performance we analyse two types of processes that can be used in the cloud-based IoT networks with different levels of parallelisation capabilities, i.e., Software-Defined Radio (SDR) and Software-Defined Networking (SDN). The results show that our agent rapidly converges to the optimal policy, stably performs in different parameter settings, outperforms or at least equally performs compared to a baseline algorithm in energy savings for different scenarios.
We study the important and challenging problem of virtualized radio access network (vRAN) design in its most general form. We develop an optimization framework that decides the number and deployment locations of central/cloud units (CUs); which distributed units (DUs) each of them will serve; the functional split that each BS will implement; and the network paths for routing the traffic to CUs and the network core. Our design criterion is to minimize the operator's expenditures while serving the expected traffic. To this end, we combine a linearization technique with a cutting-planes method in order to expedite the exact solution of the formulated problem. We evaluate our framework using real operational networks and system measurements, and follow an exhaustive parameter-sensitivity analysis. We find that the benefits when departing from single-CU deployments can be as high as 30% for our networks, but these gains diminish with the further addition of CUs. Our work sheds light on the vRAN design from a new angle, highlights the importance of deploying multiple CUs, and offers a rigorous framework for optimizing the costs of Multi-CUs vRAN.
We study the problem of wireless edge caching when file popularity is unknown and possibly non-stationary. A bank of $J$ caches receives file requests and a utility is accrued for each request depending on the serving cache. The network decides dynamically which files to store at each cache and how to route them, in order to maximize total utility. The request sequence is assumed to be drawn from an arbitrary distribution, thus capturing time-variance, temporal, or spatial locality of requests. For this challenging setting, we propose the \emph{Bipartite Supergradient Caching Algorithm} (BSCA) which provably exhibits no regret ($R_T/T \to 0$). That is, as the time horizon $T$ increases, BSCA achieves the same performance with the cache configuration that we would have chosen knowing all future requests. The learning rate of the algorithm is characterized by its regret expression, found to be $R_T=O(\sqrt{JT})$, which is independent of the content catalog size. For the single-cache case, we prove that this is the lowest attainable bound. BSCA requires at each step $J$ projections on intersections of boxes and simplices, for which we propose a tailored algorithm. Our model is the first that draws a connection between the network caching problem and Online Convex Optimization, and we demonstrate its generality by discussing various practical extensions and presenting a trace-driven comparison with state-of-the-art competitors.
Storage resources and caching techniques permeate almost every area of communication networks today. In the near future, caching is set to play an important role in storage-assisted Internet architectures, information-centric networks, and wireless systems, reducing operating and capital expenditures and improving the services offered to users. In light of the remarkable data traffic growth and the increasing number of rich-media applications, the impact of caching is expected to become even more profound than it is today. Therefore, it is crucial to design these systems in an optimal fashion, ensuring the maximum possible performance and economic benefits from their deployment. To this end, this article presents a collection of detailed models and algorithms, which are synthesized to build a powerful analytical framework for caching optimization.
This paper addresses Online Convex Optimization (OCO) problems where the constraints have additive perturbations that (i) vary over time and (ii) are not known at the time to make a decision. Perturbations may not be i.i.d. generated and can be used to model a time-varying budget or commodity in resource allocation problems. The problem is to design a policy that obtains sublinear regret while ensuring that the constraints are satisfied on average. To solve this problem, we present a primal-dual proximal gradient algorithm that has $O(T^ε\vee T^{1-ε})$ regret and $O(T^ε)$ constraint violation, where $ε\in [0,1)$ is a parameter in the learning rate. Our results match the bounds of previous work on OCO with time-varying constraints when $ε= 1/2$; however, we (i) define the regret using a time-varying set of best fixed decisions; (ii) can balance between regret and constraint violation; and (iii) use an adaptive learning rate that allows us to run the algorithm for any time horizon.