A Bayesian formulation of search, control and the exploration/exploitation trade-off

A new approach to optimisation is introduced based on a precise probabilistic statement of what is ideally required of an optimisation method. It is convenient to express the formalism in terms of the control of a stationary environment. This leads to an objective function for the controller which unifies the objectives of exploration and exploitation, thereby providing a quantitative principle for managing this trade-off. This is demonstrated using a variant of the multi-armed bandit problem. This approach opens new possibilities for optimisation algorithms, particularly by using neural network or other adaptive methods for the adaptive controller. It also opens possibilities for deepening understanding of existing methods. The realisation of these possibilities requires research into practical approximations of the exact formalism.

Stochastic control strategies and adaptive critic methods

Adaptive critic methods have common roots as generalizations of dynamic programming for neural reinforcement learning approaches. Since they approximate the dynamic programming solutions, they are potentially suitable for learning in noisy, nonlinear and nonstationary environments. In this study, a novel probabilistic dual heuristic programming (DHP) based adaptive critic controller is proposed. Distinct to current approaches, the proposed probabilistic (DHP) adaptive critic method takes uncertainties of forward model and inverse controller into consideration. Therefore, it is suitable for deterministic and stochastic control problems characterized by functional uncertainty. Theoretical development of the proposed method is validated by analytically evaluating the correct value of the cost function which satisfies the Bellman equation in a linear quadratic control problem. The target value of the critic network is then calculated and shown to be equal to the analytically derived correct value.

Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models

This paper presents some forecasting techniques for energy demand and price prediction, one day ahead. These techniques combine wavelet transform (WT) with fixed and adaptive machine learning/time series models (multi-layer perceptron (MLP), radial basis functions, linear regression, or GARCH). To create an adaptive model, we use an extended Kalman filter or particle filter to update the parameters continuously on the test set. The adaptive GARCH model is a new contribution, broadening the applicability of GARCH methods. We empirically compared two approaches of combining the WT with prediction models: multicomponent forecasts and direct forecasts. These techniques are applied to large sets of real data (both stationary and non-stationary) from the UK energy markets, so as to provide comparative results that are statistically stronger than those previously reported. The results showed that the forecasting accuracy is significantly improved by using the WT and adaptive models. The best models on the electricity demand/gas price forecast are the adaptive MLP/GARCH with the multicomponent forecast; their MSEs are 0.02314 and 0.15384 respectively.

Neural network enhanced self tuning adaptive control application for non-linear control of dynamic systems

The main theme of research of this project concerns the study of neutral networks to control uncertain and non-linear control systems. This involves the control of continuous time, discrete time, hybrid and stochastic systems with input, state or output constraints by ensuring good performances. A great part of this project is devoted to the opening of frontiers between several mathematical and engineering approaches in order to tackle complex but very common non-linear control problems. The objectives are: 1. Design and develop procedures for neutral network enhanced self-tuning adaptive non-linear control systems; 2. To design, as a general procedure, neural network generalised minimum variance self-tuning controller for non-linear dynamic plants (Integration of neural network mapping with generalised minimum variance self-tuning controller strategies); 3. To develop a software package to evaluate control system performances using Matlab, Simulink and Neural Network toolbox. An adaptive control algorithm utilising a recurrent network as a model of a partial unknown non-linear plant with unmeasurable state is proposed. Appropriately, it appears that structured recurrent neural networks can provide conveniently parameterised dynamic models for many non-linear systems for use in adaptive control. Properties of static neural networks, which enabled successful design of stable adaptive control in the state feedback case, are also identified. A survey of the existing results is presented which puts them in a systematic framework showing their relation to classical self-tuning adaptive control application of neural control to a SISO/MIMO control. Simulation results demonstrate that the self-tuning design methods may be practically applicable to a reasonably large class of unknown linear and non-linear dynamic control systems.

On waveform analysis, the adaptive task technique, and the effect of vibration on manual control skill

A re-examination of fundamental concepts and a formal structuring of the waveform analysis problem is presented in Part I. eg. the nature of frequency is examined and a novel alternative to the classical methods of detection proposed and implemented which has the advantage of speed and independence from amplitude. Waveform analysis provides the link between Parts I and II. Part II is devoted to Human Factors and the Adaptive Task Technique. The Historical, Technical and Intellectual development of the technique is traced in a review which examines the evidence of its advantages relative to non-adaptive fixed task methods of training, skill assessment and man-machine optimisation. A second review examines research evidence on the effect of vibration on manual control ability. Findings are presented in terms of percentage increment or decrement in performance relative to performance without vibration in the range 0-0.6Rms'g'. Primary task performance was found to vary by as much as 90% between tasks at the same Rms'g'. Differences in task difficulty accounted for this difference. Within tasks vibration-added-difficulty accounted for the effects of vibration intensity. Secondary tasks were found to be largely insensitive to vibration except secondaries which involved fine manual adjustment of minor controls. Three experiments are reported next in which an adaptive technique was used to measure the % task difficulty added by vertical random and sinusoidal vibration to a 'Critical Compensatory Tracking task. At vibration intensities between 0 - 0.09 Rms 'g' it was found that random vibration added (24.5 x Rms'g')/7.4 x 100% to the difficulty of the control task. An equivalence relationship between Random and Sinusoidal vibration effects was established based upon added task difficulty. Waveform Analyses which were applied to the experimental data served to validate Phase Plane analysis and uncovered the development of a control and possibly a vibration isolation strategy. The submission ends with an appraisal of subjects mentioned in the thesis title.

Practical methods of tracking of nonstationary time series applied to real-world data

In this paper, we discuss some practical implications for implementing adaptable network algorithms applied to non-stationary time series problems. Two real world data sets, containing electricity load demands and foreign exchange market prices, are used to test several different methods, ranging from linear models with fixed parameters, to non-linear models which adapt both parameters and model order on-line. Training with the extended Kalman filter, we demonstrate that the dynamic model-order increment procedure of the resource allocating RBF network (RAN) is highly sensitive to the parameters of the novelty criterion. We investigate the use of system noise for increasing the plasticity of the Kalman filter training algorithm, and discuss the consequences for on-line model order selection. The results of our experiments show that there are advantages to be gained in tracking real world non-stationary data through the use of more complex adaptive models.

Fully probabilistic control design in an adaptive critic framework

Optimal stochastic controller pushes the closed-loop behavior as close as possible to the desired one. The fully probabilistic design (FPD) uses probabilistic description of the desired closed loop and minimizes Kullback-Leibler divergence of the closed-loop description to the desired one. Practical exploitation of the fully probabilistic design control theory continues to be hindered by the computational complexities involved in numerically solving the associated stochastic dynamic programming problem. In particular very hard multivariate integration and an approximate interpolation of the involved multivariate functions. This paper proposes a new fully probabilistic contro algorithm that uses the adaptive critic methods to circumvent the need for explicitly evaluating the optimal value function, thereby dramatically reducing computational requirements. This is a main contribution of this short paper.

A quality driven framework for adaptive video streaming in mobile wireless networks

The concern over the quality of delivering video streaming services in mobile wireless networks is addressed in this work. A framework that enhances the Quality of Experience (QoE) of end users through a quality driven resource allocation scheme is proposed. To play a key role, an objective no-reference quality metric, Pause Intensity (PI), is adopted to derive a resource allocation algorithm for video streaming. The framework is examined in the context of 3GPP Long Term Evolution (LTE) systems. The requirements and structure of the proposed PI-based framework are discussed, and results are compared with existing scheduling methods on fairness, efficiency and correlation (between the required and allocated data rates). Furthermore, it is shown that the proposed framework can produce a trade-off between the three parameters through the QoE-aware resource allocation process.

Laser-induced breakdown spectroscopy quantitative analysis method via adaptive analytical line selection and relevance vector machine regression model

Abstract A new LIBS quantitative analysis method based on analytical line adaptive selection and Relevance Vector Machine (RVM) regression model is proposed. First, a scheme of adaptively selecting analytical line is put forward in order to overcome the drawback of high dependency on a priori knowledge. The candidate analytical lines are automatically selected based on the built-in characteristics of spectral lines, such as spectral intensity, wavelength and width at half height. The analytical lines which will be used as input variables of regression model are determined adaptively according to the samples for both training and testing. Second, an LIBS quantitative analysis method based on RVM is presented. The intensities of analytical lines and the elemental concentrations of certified standard samples are used to train the RVM regression model. The predicted elemental concentration analysis results will be given with a form of confidence interval of probabilistic distribution, which is helpful for evaluating the uncertainness contained in the measured spectra. Chromium concentration analysis experiments of 23 certified standard high-alloy steel samples have been carried out. The multiple correlation coefficient of the prediction was up to 98.85%, and the average relative error of the prediction was 4.01%. The experiment results showed that the proposed LIBS quantitative analysis method achieved better prediction accuracy and better modeling robustness compared with the methods based on partial least squares regression, artificial neural network and standard support vector machine.

Adaptive resource allocation for QoE-aware mobile communication networks

A real-time adaptive resource allocation algorithm considering the end user's Quality of Experience (QoE) in the context of video streaming service is presented in this work. An objective no-reference quality metric, namely Pause Intensity (PI), is used to control the priority of resource allocation to users during the scheduling process. An online adjustment has been introduced to adaptively set the scheduler's parameter and maintain a desired trade-off between fairness and efficiency. The correlation between the data rates (i.e. video code rates) demanded by users and the data rates allocated by the scheduler is taken into account as well. The final allocated rates are determined based on the channel status, the distribution of PI values among users, and the scheduling policy adopted. Furthermore, since the user's capability varies as the environment conditions change, the rate adaptation mechanism for video streaming is considered and its interaction with the scheduling process under the same PI metric is studied. The feasibility of implementing this algorithm is examined and the result is compared with the most commonly existing scheduling methods.

Quality-driven resource utilization methods for video streaming in wireless communication networks

This research is focused on the optimisation of resource utilisation in wireless mobile networks with the consideration of the users’ experienced quality of video streaming services. The study specifically considers the new generation of mobile communication networks, i.e. 4G-LTE, as the main research context. The background study provides an overview of the main properties of the relevant technologies investigated. These include video streaming protocols and networks, video service quality assessment methods, the infrastructure and related functionalities of LTE, and resource allocation algorithms in mobile communication systems. A mathematical model based on an objective and no-reference quality assessment metric for video streaming, namely Pause Intensity, is developed in this work for the evaluation of the continuity of streaming services. The analytical model is verified by extensive simulation and subjective testing on the joint impairment effects of the pause duration and pause frequency. Various types of the video contents and different levels of the impairments have been used in the process of validation tests. It has been shown that Pause Intensity is closely correlated with the subjective quality measurement in terms of the Mean Opinion Score and this correlation property is content independent. Based on the Pause Intensity metric, an optimised resource allocation approach is proposed for the given user requirements, communication system specifications and network performances. This approach concerns both system efficiency and fairness when establishing appropriate resource allocation algorithms, together with the consideration of the correlation between the required and allocated data rates per user. Pause Intensity plays a key role here, representing the required level of Quality of Experience (QoE) to ensure the best balance between system efficiency and fairness. The 3GPP Long Term Evolution (LTE) system is used as the main application environment where the proposed research framework is examined and the results are compared with existing scheduling methods on the achievable fairness, efficiency and correlation. Adaptive video streaming technologies are also investigated and combined with our initiatives on determining the distribution of QoE performance across the network. The resulting scheduling process is controlled through the prioritization of users by considering their perceived quality for the services received. Meanwhile, a trade-off between fairness and efficiency is maintained through an online adjustment of the scheduler’s parameters. Furthermore, Pause Intensity is applied to act as a regulator to realise the rate adaptation function during the end user’s playback of the adaptive streaming service. The adaptive rates under various channel conditions and the shape of the QoE distribution amongst the users for different scheduling policies have been demonstrated in the context of LTE. Finally, the work for interworking between mobile communication system at the macro-cell level and the different deployments of WiFi technologies throughout the macro-cell is presented. A QoEdriven approach is proposed to analyse the offloading mechanism of the user’s data (e.g. video traffic) while the new rate distribution algorithm reshapes the network capacity across the macrocell. The scheduling policy derived is used to regulate the performance of the resource allocation across the fair-efficient spectrum. The associated offloading mechanism can properly control the number of the users within the coverages of the macro-cell base station and each of the WiFi access points involved. The performance of the non-seamless and user-controlled mobile traffic offloading (through the mobile WiFi devices) has been evaluated and compared with that of the standard operator-controlled WiFi hotspots.

Self-adaptive probabilistic roadmap generation for intelligent virtual agents

Agents inhabiting large scale environments are faced with the problem of generating maps by which they can navigate. One solution to this problem is to use probabilistic roadmaps which rely on selecting and connecting a set of points that describe the interconnectivity of free space. However, the time required to generate these maps can be prohibitive, and agents do not typically know the environment in advance. In this paper we show that the optimal combination of different point selection methods used to create the map is dependent on the environment, no point selection method dominates. This motivates a novel self-adaptive approach for an agent to combine several point selection methods. The success rate of our approach is comparable to the state of the art and the generation cost is substantially reduced. Self-adaptation therefore enables a more efficient use of the agent's resources. Results are presented for both a set of archetypal scenarios and large scale virtual environments based in Second Life, representing real locations in London.

Temperature management for heterogeneous multi-core FPGAs using adaptive evolutionary multi-objective approaches

Heterogeneous multi-core FPGAs contain different types of cores, which can improve efficiency when used with an effective online task scheduler. However, it is not easy to find the right cores for tasks when there are multiple objectives or dozens of cores. Inappropriate scheduling may cause hot spots which decrease the reliability of the chip. Given that, our research builds a simulating platform to evaluate all kinds of scheduling algorithms on a variety of architectures. On this platform, we provide an online scheduler which uses multi-objective evolutionary algorithm (EA). Comparing the EA and current algorithms such as Predictive Dynamic Thermal Management (PDTM) and Adaptive Temperature Threshold Dynamic Thermal Management (ATDTM), we find some drawbacks in previous work. First, current algorithms are overly dependent on manually set constant parameters. Second, those algorithms neglect optimization for heterogeneous architectures. Third, they use single-objective methods, or use linear weighting method to convert a multi-objective optimization into a single-objective optimization. Unlike other algorithms, the EA is adaptive and does not require resetting parameters when workloads switch from one to another. EAs also improve performance when used on heterogeneous architecture. A efficient Pareto front can be obtained with EAs for the purpose of multiple objectives.

Formal methods @ runtime

Heuristics, simulation, artificial intelligence techniques and combinations thereof have all been employed in the attempt to make computer systems adaptive, context-aware, reconfigurable and self-managing. This paper complements such efforts by exploring the possibility to achieve runtime adaptiveness using mathematically-based techniques from the area of formal methods. It is argued that formal methods @ runtime represents a feasible approach, and promising preliminary results are summarised to support this viewpoint. The survey of existing approaches to employing formal methods at runtime is accompanied by a discussion of their challenges and of the future research required to overcome them. © 2011 Springer-Verlag.