A novel adaptive weight selection algorithm for multi-objective multi-agent reinforcement learning

To solve multi-objective problems, multiple reward signals are often scalarized into a single value and further processed using established single-objective problem solving techniques. While the field of multi-objective optimization has made many advances in applying scalarization techniques to obtain good solution trade-offs, the utility of applying these techniques in the multi-objective multi-agent learning domain has not yet been thoroughly investigated. Agents learn the value of their decisions by linearly scalarizing their reward signals at the local level, while acceptable system wide behaviour results. However, the non-linear relationship between weighting parameters of the scalarization function and the learned policy makes the discovery of system wide trade-offs time consuming. Our first contribution is a thorough analysis of well known scalarization schemes within the multi-objective multi-agent reinforcement learning setup. The analysed approaches intelligently explore the weight-space in order to find a wider range of system trade-offs. In our second contribution, we propose a novel adaptive weight algorithm which interacts with the underlying local multi-objective solvers and allows for a better coverage of the Pareto front. Our third contribution is the experimental validation of our approach by learning bi-objective policies in self-organising smart camera networks. We note that our algorithm (i) explores the objective space faster on many problem instances, (ii) obtained solutions that exhibit a larger hypervolume, while (iii) acquiring a greater spread in the objective space.

Exposing market mechanism design trade-offs via multi-objective evolutionary search

Market mechanisms are a means by which resources in contention can be allocated between contending parties, both in human economies and those populated by software agents. Designing such mechanisms has traditionally been carried out by hand, and more recently by automation. Assessing these mechanisms typically involves them being evaluated with respect to multiple conflicting objectives, which can often be nonlinear, noisy, and expensive to compute. For typical performance objectives, it is known that designed mechanisms often fall short on being optimal across all objectives simultaneously. However, in all previous automated approaches, either only a single objective is considered, or else the multiple performance objectives are combined into a single objective. In this paper we do not aggregate objectives, instead considering a direct, novel application of multi-objective evolutionary algorithms (MOEAs) to the problem of automated mechanism design. This allows the automatic discovery of trade-offs that such objectives impose on mechanisms. We pose the problem of mechanism design, specifically for the class of linear redistribution mechanisms, as a naturally existing multi-objective optimisation problem. We apply a modified version of NSGA-II in order to design mechanisms within this class, given economically relevant objectives such as welfare and fairness. This application of NSGA-II exposes tradeoffs between objectives, revealing relationships between them that were otherwise unknown for this mechanism class. The understanding of the trade-off gained from the application of MOEAs can thus help practitioners with an insightful application of discovered mechanisms in their respective real/artificial markets.

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.

Towards understanding IT value co-creation in crowdsourcing:the multiple stakeholders' perspective research-in-progress paper

Contemporary business environment involves IT being invested and shared by multiple stakeholders in collaborative, platform-based, and relational arrangements where the objective is to co-create value. Traditional IT enabled business value therefore has been extended towards IT value co-creation that involves multiple stakeholders. In this paper, we present a conceptual development of IT-based value co-creation in the context of online crowdsourcing. Based on the existing literature, we have distinguished multiple crowdsourcing types (models) by analyzing attributes of crowd, the roles of the client, the platform and the crowd that act as key stakeholders in the value co-creation process, and describe the major interactions between the main stakeholders. Our conceptual development is suggesting different combinations of value co-creation layers to be evident in different crowdsourcing models.

Understanding and improving the manufacturing and changeover process in Saudi Arabian business - a multiple case study approach

The importance of the changeover process in the manufacturing industry is becoming widely recognised. Changeover is a complete process of changing between the manufacture of one product to manufacture of an alternative product until specified production and quality rates are reached. The initiatives to improve changeover exist in industry, as better changeover process typically contribute to improved quality performance. A high-quality and reliable changeover process can be achieved through implementation of continuous or radical improvements. This research examines the changeover process of Saudi Arabian manufacturing firms because Saudi Arabia’s government is focused on the expansion of GDP and increasing the number of export manufacturing firms. Furthermore, it is encouraging foreign manufacturing firms to invest within Saudi Arabia. These initiatives, therefore, require that Saudi manufacturing businesses develop the changeover practice in order to compete in the market and achieve the government’s objectives. Therefore, the aim of this research is to discover the current status of changeover process implementation in Saudi Arabian manufacturing businesses. To achieve this aim, the main objective of this research is to develop a conceptual model to understand and examine the effectiveness of the changeover process within Saudi Arabian manufacturing firms, facilitating identification of those activities that affect the reliability and high-quality of the process. In order to provide a comprehensive understanding of this area, this research first explores the concept of quality management and its relationship to firm performance and the performance of manufacturing changeover. An extensive body of literature was reviewed on the subject of lean manufacturing and changeover practice. A research conceptual model was identified based on this review, and focus was on providing high-quality and reliable manufacturing changeover processes during set-up in a dynamic environment. Exploratory research was conducted in sample Saudi manufacturing firms to understand the features of the changeover process within the manufacturing sector, and as a basis for modifying the proposed conceptual model. Qualitative research was employed in the study with semi-structured interviews, direct observations and documentation in order to understand the real situation such as actual daily practice and current status of changeover process in the field. The research instrument, the Changeover Effectiveness Assessment Tool (CEAT) was developed to evaluate changeover practices. A pilot study was conducted by examining the CEAT, proposed for the main research. Consequently, the conceptual model was modified and CEAT was improved in response to the pilot study findings. Case studies have been conducted within eight Saudi manufacturing businesses. These case studies assessed the implementation of manufacturing changeover practice in the lighting and medical products sectors. These two sectors were selected based on their operation strategy which was batch production as well as the fact that they fulfilled the research sampling strategy. The outcomes of the research improved the conceptual model, ultimately to facilitate the firms’ adoption and rapid implementation of a high-quality and reliability changeover during the set-up process. The main finding of this research is that Quality’s factors were considering the lowest levels comparing to the other factors which are People, Process and Infrastructure. This research contributes to enable Saudi businesses to implement the changeover process by adopting the conceptual model. In addition, the guidelines for facilitating implementation were provided in this thesis. Therefore, this research provides insight to enable the Saudi manufacturing industry to be more responsive to rapidly changing customer demands.