Tuning compilations by multi-objective optimization: Application to Apache web server

Modern compilers present a great and ever increasing number of options which can modify the features and behavior of a compiled program. Many of these options are often wasted due to the required comprehensive knowledge about both the underlying architecture and the internal processes of the compiler. In this context, it is usual, not having a single design goal but a more complex set of objectives. In addition, the dependencies between different goals are difficult to be a priori inferred. This paper proposes a strategy for tuning the compilation of any given application. This is accomplished by using an automatic variation of the compilation options by means of multi-objective optimization and evolutionary computation commanded by the NSGA-II algorithm. This allows finding compilation options that simultaneously optimize different objectives. The advantages of our proposal are illustrated by means of a case study based on the well-known Apache web server. Our strategy has demonstrated an ability to find improvements up to 7.5% and up to 27% in context switches and L2 cache misses, respectively, and also discovers the most important bottlenecks involved in the application performance.

Feature selection by multi-objective optimisation: Application to network anomaly detection by hierarchical self-organising maps

Feature selection is an important and active issue in clustering and classification problems. By choosing an adequate feature subset, a dataset dimensionality reduction is allowed, thus contributing to decreasing the classification computational complexity, and to improving the classifier performance by avoiding redundant or irrelevant features. Although feature selection can be formally defined as an optimisation problem with only one objective, that is, the classification accuracy obtained by using the selected feature subset, in recent years, some multi-objective approaches to this problem have been proposed. These either select features that not only improve the classification accuracy, but also the generalisation capability in case of supervised classifiers, or counterbalance the bias toward lower or higher numbers of features that present some methods used to validate the clustering/classification in case of unsupervised classifiers. The main contribution of this paper is a multi-objective approach for feature selection and its application to an unsupervised clustering procedure based on Growing Hierarchical Self-Organising Maps (GHSOMs) that includes a new method for unit labelling and efficient determination of the winning unit. In the network anomaly detection problem here considered, this multi-objective approach makes it possible not only to differentiate between normal and anomalous traffic but also among different anomalies. The efficiency of our proposals has been evaluated by using the well-known DARPA/NSL-KDD datasets that contain extracted features and labelled attacks from around 2 million connections. The selected feature sets computed in our experiments provide detection rates up to 99.8% with normal traffic and up to 99.6% with anomalous traffic, as well as accuracy values up to 99.12%.

Robust solutions to multi-objective linear programs with uncertain data

In this paper we examine multi-objective linear programming problems in the face of data uncertainty both in the objective function and the constraints. First, we derive a formula for the radius of robust feasibility guaranteeing constraint feasibility for all possible scenarios within a specified uncertainty set under affine data parametrization. We then present numerically tractable optimality conditions for minmax robust weakly efficient solutions, i.e., the weakly efficient solutions of the robust counterpart. We also consider highly robust weakly efficient solutions, i.e., robust feasible solutions which are weakly efficient for any possible instance of the objective matrix within a specified uncertainty set, providing lower bounds for the radius of highly robust efficiency guaranteeing the existence of this type of solutions under affine and rank-1 objective data uncertainty. Finally, we provide numerically tractable optimality conditions for highly robust weakly efficient solutions.

Intelligent Decision-Making System Frameworks for a Digital Plan of Work: A Theoretical Investigation for the Construction Industry

The construction industry is characterised by fragmentation and suffers from lack of collaboration, often adopting adversarial working practices to achieve deliverables. For the UK Government and construction industry, BIM is a game changer aiming to rectify this fragmentation and promote collaboration. However it has become clear that there is an essential need to have better controls and definitions of both data deliverables and data classification. Traditional methods and techniques for collating and inputting data have shown to be time consuming and provide little to improve or add value to the overall task of improving deliverables. Hence arose the need in the industry to develop a Digital Plan of Work (DPoW) toolkit that would aid the decision making process, providing the required control over the project workflows and data deliverables, and enabling better collaboration through transparency of need and delivery. The specification for the existing Digital Plan of Work (DPoW) was to be, an industry standard method of describing geometric, requirements and data deliveries at key stages of the project cycle, with the addition of a structured and standardised information classification system. However surveys and interviews conducted within this research indicate that the current DPoW resembles a digitised version of the pre-existing plans of work and does not push towards the data enriched decision-making abilities that advancements in technology now offer. A Digital Framework is not simply the digitisation of current or historic standard methods and procedures, it is a new intelligent driven digital system that uses new tools, processes, procedures and work flows to eradicate waste and increase efficiency. In addition to reporting on conducted surveys above, this research paper will present a theoretical investigation into usage of Intelligent Decision Support Systems within a digital plan of work framework. Furthermore this paper will present findings on the suitability to utilise advancements in intelligent decision-making system frameworks and Artificial Intelligence for a UK BIM Framework. This should form the foundations of decision-making for projects implemented at BIM level 2. The gap identified in this paper is that the current digital toolkit does not incorporate the intelligent characteristics available in other industries through advancements in technology and collation of vast amounts of data that a digital plan of work framework could have access to and begin to develop, learn and adapt for decision-making through the live interaction of project stakeholders.

Sport-Specific Practice and the Development of Expert Decision-Making in Team Ball Sports

The role of sport-specific practice in the development of decision-making expertise in the sports of field hockey, netball, and basketball was examined. Fifteen expert decision-makers and 13 experienced non-expert athletes provided detailed information about the quantity and type of sport-specific and other related practice activities they had undertaken throughout their careers. Experts accumulated more hours of sport-specific practice from age 12 years onwards than did non-experts, spending on average some 13 years and 4,000 hours on concentrated sport-specific practice before reaching international standard. A significant negative correlation existed between the number of additional activities undertaken and the hours of sportspecific training required before attaining expertise, suggesting a functional role for activities other than sport-specific training in the development of expert decision-making.

The Contribution of Structured Activity and Deliberate Play to the Development of Expert Perceptual and Decision-Making Skill

The developmental histories of 32 players in the Australian Football League (AFL), independently classified as either expert or less skilled in their perceptual and decision- making skills, were collected through a structured interview process and their year-on-year involvement in structured and deliberate play activities retrospectively determined. Despite being drawn from the same elite level of competition, the expert decision-makers differed from the less skilled in having accrued, during their developing years, more hours of experience in structured activities of all types, in structured activities in invasion-type sports, in invasion-type deliberate play, and in invasion activities from sports other than Australian football. Accumulated hours invested in invasion-type activities differentiated between the groups, suggesting that it is the amount of invasion-type activity that is experienced and not necessarily intent (skill development or fun) or specificity that facilitates the development of perceptual and decision-making expertise in this team sport.

The Optimality of Decision Making During Motor Learning

In our daily lives, we often must predict how well we are going to perform in the future based on an evaluation of our current performance and an assessment of how much we will improve with practice. Such predictions can be used to decide whether to invest our time and energy in learning and, if we opt to invest, what rewards we may gain. This thesis investigated whether people are capable of tracking their own learning (i.e. current and future motor ability) and exploiting that information to make decisions related to task reward. In experiment one, participants performed a target aiming task under a visuomotor rotation such that they initially missed the target but gradually improved. After briefly practicing the task, they were asked to select rewards for hits and misses applied to subsequent performance in the task, where selecting a higher reward for hits came at a cost of receiving a lower reward for misses. We found that participants made decisions that were in the direction of optimal and therefore demonstrated knowledge of future task performance. In experiment two, participants learned a novel target aiming task in which they were rewarded for target hits. Every five trials, they could choose a target size which varied inversely with reward value. Although participants’ decisions deviated from optimal, a model suggested that they took into account both past performance, and predicted future performance, when making their decisions. Together, these experiments suggest that people are capable of tracking their own learning and using that information to make sensible decisions related to reward maximization.

Bridging qualified majority and unanimity decision-making in the EU. IHS Political Science Series No. 132, February 2013

The EU has tried to bridge decision making by qualified majority and unanimity over the years by expanding qualified majorities (consensus) or by making unanimities easier to achieve. I call this decision-making procedure q-“unanimity” and trace its history from the Luxembourg compromise to the Lisbon Treaty, and to more recent agreements. I analyze the most recent and explicit mechanism of this bridging (article 31 (2) of the Lisbon Treaty) and identify one specific means by which the transformation of qualified majorities to unanimities is achieved: the reduction of precision or scope of the decision, so that different behaviors can be covered by it. I provide empirical evidence of such a mechanism by analyzing legislative decisions. Finally, I argue that this bridging is a ubiquitous feature of EU institutions, used in Treaties as well as in legislative decision-making.

Improving decision-making in the EU. CHALLENGE EUROPE Issue 22 - Challenges and new beginnings: Priorities for the EU's new leadership, September 2014.

The multiple crises the European Union (EU) has experienced in recent years have fundamentally altered decision-making and, more broadly, governance in the EU. Pre-crisis systems and processes were not adequate to react to such critical and systemic challenges, but the speed of the crisis meant that new governance mechanisms have been superimposed on existing processes and structures rather than seeing a fundamental reform of decision-making. Consequently, not all changes have been fully successful. Given the institutional changes this year and the ongoing development of the EMU governance framework, now presents a good opportunity to reform EU decision-making.