Electrical and Electronic Engineering - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 19
-
ItemNo Preview AvailableAn Information-Theoretic Analysis for Transfer Learning: Error Bounds and Applications( 2022-07-12)Transfer learning, or domain adaptation, is concerned with machine learning problems in which training and testing data come from possibly different probability distributions. In this work, we give an information-theoretic analysis on the generalization error and excess risk of transfer learning algorithms, following a line of work initiated by Russo and Xu. Our results suggest, perhaps as expected, that the Kullback-Leibler (KL) divergence D(μ||μ′) plays an important role in the characterizations where μ and μ′ denote the distribution of the training data and the testing test, respectively. Specifically, we provide generalization error upper bounds for the empirical risk minimization (ERM) algorithm where data from both distributions are available in the training phase. We further apply the analysis to approximated ERM methods such as the Gibbs algorithm and the stochastic gradient descent method. We then generalize the mutual information bound with ϕ-divergence and Wasserstein distance. These generalizations lead to tighter bounds and can handle the case when μ is not absolutely continuous with respect to μ′. Furthermore, we apply a new set of techniques to obtain an alternative upper bound which gives a fast (and optimal) learning rate for some learning problems. Finally, inspired by the derived bounds, we propose the InfoBoost algorithm in which the importance weights for source and target data are adjusted adaptively in accordance to information measures. The empirical results show the effectiveness of the proposed algorithm.
-
ItemAchieving QoS for Real-Time Bursty Applications over Passive Optical Networks( 2021-09-05)Emerging real-time applications such as those classified under ultra-reliable low latency (uRLLC) generate bursty traffic and have strict Quality of Service (QoS) requirements. Passive Optical Network (PON) is a popular access network technology, which is envisioned to handle such applications at the access segment of the network. However, the existing standards cannot handle strict QoS constraints. The available solutions rely on instantaneous heuristic decisions and maintain QoS constraints (mostly bandwidth) in an average sense. Existing works with optimal strategies are computationally complex and are not suitable for uRLLC applications. This paper presents a novel computationally-efficient, far-sighted bandwidth allocation policy design for facilitating bursty traffic in a PON framework while satisfying strict QoS (age of information/delay and bandwidth) requirements of modern applications. To this purpose, first we design a delay-tracking mechanism which allows us to model the resource allocation problem from a control-theoretic viewpoint as a Model Predictive Control (MPC). MPC helps in taking far-sighted decisions regarding resource allocations and captures the time-varying dynamics of the network. We provide computationally efficient polynomial-time solutions and show its implementation in the PON framework. Compared to existing approaches, MPC reduces delay violations by approximately 15% for a delay-constrained application of 1ms target. Our approach is also robust to varying traffic arrivals.
-
ItemAchieving AI-enabled Robust End-to-End Quality of Experience over Radio Access Networks( 2022-01-13)Emerging applications such as Augmented Reality, the Internet of Vehicles and Remote Surgery require both computing and networking functions working in harmony. The End-to-end (E2E) quality of experience (QoE) for these applications depends on the synchronous allocation of networking and computing resources. However, the relationship between the resources and the E2E QoE outcomes is typically stochastic and non-linear. In order to make efficient resource allocation decisions, it is essential to model these relationships. This article presents a novel machine-learning based approach to learn these relationships and concurrently orchestrate both resources for this purpose. The machine learning models further help make robust allocation decisions regarding stochastic variations and simplify robust optimization to a conventional constrained optimization. When resources are insufficient to accommodate all application requirements, our framework supports executing some of the applications with minimal degradation (graceful degradation) of E2E QoE. We also show how we can implement the learning and optimization methods in a distributed fashion by the Software-Defined Network (SDN) and Kubernetes technologies. Our results show that deep learning-based modelling achieves E2E QoE with approximately 99.8\% accuracy, and our robust joint-optimization technique allocates resources efficiently when compared to existing differential services alternatives.
-
Item
-
ItemWhen to stop value iteration: stability and near-optimality versus computation( 2020-11-19)Value iteration (VI) is a ubiquitous algorithm for optimal control, planning, and reinforcement learning schemes. Under the right assumptions, VI is a vital tool to generate inputs with desirable properties for the controlled system, like optimality and Lyapunov stability. As VI usually requires an infinite number of iterations to solve general nonlinear optimal control problems, a key question is when to terminate the algorithm to produce a “good” solution, with a measurable impact on optimality and stability guarantees. By carefully analysing VI under general stabilizability and detectability properties, we provide explicit and novel relationships of the stopping criterion’s impact on near-optimality, stability and performance, thus allowing to tune these desirable properties against the induced computational cost. The considered class of stopping criteria encompasses those encountered in the control, dynamic programming and reinforcement learning literature and it allows considering new ones, which may be useful to further reduce the computational cost while endowing and satisfying stability and near-optimality properties. We therefore lay a foundation to endow machine learning schemes based on VI with stability and performance guarantees, while reducing computational complexity.
-
ItemActive Learning for Linear Parameter-Varying System Identification( 2020-05-02)Active learning is proposed for selection of the next operating points in the design of experiments, for identifying linear parameter-varying systems. We extend existing approaches found in literature to multiple-input multiple-output systems with a multivariate scheduling parameter. Our approach is based on exploiting the probabilistic features of Gaussian process regression to quantify the overall model uncertainty across locally identified models. This results in a flexible framework which accommodates for various techniques to be applied for estimation of local linear models and their corresponding uncertainty. We perform active learning in application to the identification of a diesel engine air-path model, and demonstrate that measures of model uncertainty can be successfully reduced using the proposed framework.
-
ItemTracking and regret bounds for online zeroth-order Euclidean and Riemannian optimisation( 2020-10-01)We study numerical optimisation algorithms that use zeroth-order information to minimise time-varying geodesically-convex cost functions on Riemannian manifolds. In the Euclidean setting, zeroth-order algorithms have received a lot of attention in both the time-varying and time-invariant cases. However, the extension to Riemannian manifolds is much less developed. We focus on Hadamard manifolds, which are a special class of Riemannian manifolds with global nonpositive curvature that offer convenient grounds for the generalisation of convexity notions. Specifically, we derive bounds on the expected instantaneous tracking error, and we provide algorithm parameter values that minimise the algorithm’s performance. Our results illustrate how the manifold geometry in terms of the sectional curvature affects these bounds. Additionally, we provide dynamic regret bounds for this online optimisation setting. To the best of our knowledge, these are the first regret bounds even for the Euclidean version of the problem. Lastly, via numerical simulations, we demonstrate the applicability of our algorithm on an online Karcher mean problem.
-
ItemOn the Latency, Rate and Reliability Tradeoff in Wireless Networked Control Systems for IIoT( 2020-07-01)Wireless networked control systems (WNCSs) provide a key enabling technique for Industry Internet of Things (IIoT). However, in the literature of WNCSs, most of the research focuses on the control perspective, and has considered oversimplified models of wireless communications which do not capture the key parameters of a practical wireless communication system, such as latency, data rate and reliability. In this paper, we focus on a WNCS, where a controller transmits quantized and encoded control codewords to a remote actuator through a wireless channel, and adopt a detailed model of the wireless communication system, which jointly considers the inter-related communication parameters. We derive the stability region of the WNCS. If and only if the tuple of the communication parameters lies in the region, the average cost function, i.e., a performance metric of the WNCS, is bounded. We further obtain a necessary and sufficient condition under which the stability region is n-bounded, where n is the control codeword blocklength. We also analyze the average cost function of the WNCS. Such analysis is non-trivial because the finite-bit control-signal quantizer introduces a non-linear and discontinuous quantization function which makes the performance analysis very difficult. We derive tight upper and lower bounds on the average cost function in terms of latency, data rate and reliability. Our analytical results provide important insights into the design of the optimal parameters to minimize the average cost within the stability region.
-
ItemOn Joint Reconstruction of State and Input-Output Injection Attacks for Nonlinear Systems( 2021-03-08)We address the problem of robust state reconstruction for discrete-time nonlinear systems when the actuators and sensors are injected with (potentially unbounded) attack signals. Exploiting redundancy in sensors and actuators and using a bank of unknown input observers (UIOs), we propose an observer-based estimator capable of providing asymptotic estimates of the system state and attack signals under the condition that the numbers of sensors and actuators under attack are sufficiently small. Using the proposed estimator, we provide methods for isolating the compromised actuators and sensors. Numerical examples are provided to demonstrate the effectiveness of our methods.
-
ItemA Robust CACC Scheme Against Cyberattacks Via Multiple Vehicle-to-Vehicle Networks( 2021-06-19)Cooperative Adaptive Cruise Control (CACC) is a vehicular technology that allows groups of vehicles on the highway to form in closely-coupled automated platoons to increase highway capacity and safety, and decrease fuel consumption and CO2 emissions. The underlying mechanism behind CACC is the use of Vehicle-to-Vehicle (V2V) wireless communication networks to transmit acceleration commands to adjacent vehicles in the platoon. However, the use of V2V networks leads to increased vulnerabilities against faults and cyberattacks at the communication channels. Communication networks serve as new access points for malicious agents trying to deteriorate the platooning performance or even cause crashes. Here, we address the problem of increasing robustness of CACC schemes against cyberattacks by the use of multiple V2V networks and a data fusion algorithm. The idea is to transmit acceleration commands multiple times through different communication networks (channels) to create redundancy at the receiver side. We exploit this redundancy to obtain attack-free estimates of acceleration commands. To accomplish this, we propose a data-fusion algorithm that takes data from all channels, returns an estimate of the true acceleration command, and isolates compromised channels. Note, however, that using estimated data for control introduces uncertainty into the loop and thus decreases performance. To minimize performance degradation, we propose a robust H∞ controller that reduces the joint effect of estimation errors and sensor/channel noise in the platooning performance (tracking performance and string stability). We present simulation results to illustrate the performance of our approach.