Interpretable machine learning is important for scientific, biomedical, and other high-stakes domains where transparency, trust, and human understanding of models are required. Recently, an evolutionary rule-based machine learning algorithm called HEROS was proposed, which separated rule from rule-set discovery each with distinct multi-objective optimization strategies yielding a two-phase, noise agnostic framework that found accurate, highly compact, and interpretable solutions on a diverse set of challenging benchmarks without requiring hyperparameter optimizations. However, HEROS is currently limited by its strictly sequential two-phase design which requires committing in advance to a fixed number of rule-discovery iterations. This work extends HEROS with a phase alternation scheme interleaving rule discovery and rule-set optimization. This enables refinement of both tasks throughout training in an 'anytime-algorithm' manner. We also introduce a tree-based rule initialization strategy to accelerate early-stage rule discovery. These extensions aim to preserve the empirical performance of the original HEROS framework under a similar computational budget while providing greater flexibility for algorithm application and problem scalability. Using the same diverse set of benchmark datasets, we evaluate and compare these HEROS extensions to the original HEROS algorithm and established rule-based learning systems including RIPPER and BioHEL. We demonstrate the advantages of this extended HEROS framework.

• Two scenarios where smart production planning assistance improves human workflows • A questionnaire to assess levels of explainability needed to implement this system • Survey results by a group of active production planners (n=11) • Light levels of XAI are a strict requirement for an AI-based smart production system • Higher explainability is requested but should not harm performance In recent years, intelligent systems have increased their capabilities greatly increasing their practical applicability. However, for the foreseeable future, such AI-powered agents will not act autonomously but assist a human that will ultimately responsible. Here, the explainability of agents’ suggestions becomes paramount to provide trust and acceptance by their human co-workers. For the field of stochastic/evolutionary optimization, it has not yet been investigated what levels of explainability real human stakeholders without deep technical knowledge of these systems actually request. In this article, we report an exploratory case study where we questioned a group of production planners ( n = 11 ) about their needs for AI assistance and what types of explanations they would require to integrate AI into their day-to-day work-flow and still feel comfortable with the cooperation. While five participants expect their individual position to be threatened by these systems in the mid-term, all participants agree that AI is beneficial for safeguarding the location against competitors or migration. We find that AI-based assistance is requested to a large degree across all age groups and that stakeholders greatly request explainability of the agent’s recommendations. From this real-world empirical evidence it becomes evident that implementing explainable optimization in production planning is a crucial next step towards Industry 5.0.

Rule-based machine learning (RBML) models are often presumed to be very beneficial for tasks where explainabil-ity of machine learning models is considered essential. However, their models are only really explainable as long as their rule sets are compact. This leads to the need for an optimizer to take prediction error and rule count as objectives. Given the highly complex fitness landscape of rule set learning tasks, good hyperparameters of the optimizer as well as their robustness against local minima is detrimental. In this paper, we explore the use of four self-adaptive genetic algorithms (SAGAs) for the optimization of a recent evolutionary RBML system to reduce the number of hyperparameters to tune and hopefully find better minima. To evaluate the advantages, we benchmark against a non-adaptive genetic algorithm (GA) on five real-world data sets. We find-with the support of a rigorous statistical analysis-that some of the SAGAs deliver a suitable alternative, which is easier to handle for non-experts in GA configurations. This is crucial for a wider application of this RBML method.

Practitioners frequently encounter the challenge of selecting the best optimization algorithm from a pool of options. However, why not, rather than selecting a single algorithm, let evolution determine the optimal combination of all algorithms? In this paper, we present an approach to algorithm design inspired by a well-known traditional method for coarse-grained hybridization: the heterogeneous island model. Our hyper-heuristic framework represents island models as graphs and identifies optimal island topologies and parameters for specific sets of problem instances. Since the framework operates at the level of metaheuristic algorithms rather than components and incorporates a configuration mechanism directly into the search, it combines concepts from algorithm design, selection, and configuration. The proposed framework is investigated on 24 training sets of varying difficulty and demonstrates its ability to discover complex hybrids. A post-evaluation on real-world constrained optimization problems shows a significant improvement over the algorithms on their own. These results suggest that it is a promising way to design hybrid metaheuristics with minimal manual intervention, given representative training instances, a set of optimization algorithms, and sufficient computational resources.

The Machine Health (MH) sector—which includes, for example, Predictive Maintenance, Prognostics and Health Management, and Condition Monitoring—has the potential to improve efficiency and reduce costs for maintenance and machine operation. This is achieved by data-driven analytics applications, utilising the vast amount of data collected by sensors during machine runtime. While there are numerous possible fields of application, the overall complexity of machines and applications in scientific publications is still low, preventing MH technologies from being implemented in many real-world scenarios. This may be the result of a diffuse understanding of the term complexity in the publications of this field, which results in a lack of focus towards the core problems of real-world MH applications. This article introduces a new way of discerning complexity in data-driven MH applications, enabling an effective discussion and analysis of present and future MH applications. This is achieved by creating a new taxonomy based on observations from relevant literature and substantial domain knowledge. Using this newly introduced taxonomy, we categorise recent applications of MH to demonstrate the usefulness of our approach and illustrate a still-prevalent research gap based on our findings.

Maintenance of complex machinery is time and resource intensive. Therefore, decreasing maintenance cycles by employing Predictive Maintenance (PdM) is sought after by many manufacturers of machines and can be a valuable selling point. However, currently PdM is a hard to solve problem getting increasingly harder with the complexity of the maintained system. One challenge is to adequately prepare data for model training and analysis. In this paper, we propose the use of expert knowledge–based preprocessing techniques to extend the standard data science–workflow. We define complex multi-purpose machinery as an application domain and test our proposed techniques on real-world data generated by numerous machines deployed in the wild. We find that our techniques enable and enhance model training.

308

Analysing the behaviour of metaheuristics comprehensively and thereby enhancing explainability requires large empirical studies. However, the amount of data gathered in such experiments is often too large to be examined and evaluated visually. This necessitates establishing more efficient analysis procedures, but care has to be taken so that these do not obscure important information. This paper examines the suitability of clustering methods to assist in the characterisation of the behaviour of metaheuristics. The convergence behaviour is used as an example as its empirical analysis often requires looking at convergence curve plots, which is extremely tedious for large algorithmic datasets. We used the well-known K-Means clustering method and examined the results for different cluster sizes. Furthermore, we evaluated the clusters with respect to the characteristics they utilise and compared those with characteristics applied when a researcher inspects convergence curve plots. We found that clustering is a suitable technique to assist in the analysis of convergence behaviour, as the clusters strongly correspond to the grouping that would be done by a researcher, though the procedure still requires background knowledge to determine an adequate number of clusters. Overall, this enables us to inspect only few curves per cluster instead of all individual curves.

In sociotechnical settings, human operators are increasingly assisted by decision support systems. By employing such systems, important properties of sociotechnical systems, such as self-adaptation and self-optimization, are expected to improve further. To be accepted by and engage efficiently with operators, decision support systems need to be able to provide explanations regarding the reasoning behind specific decisions. In this article, we propose the use of learning classifier systems (LCSs), a family of rule-based machine learning methods, to facilitate and highlight techniques to improve transparent decision-making. Furthermore, we present a novel approach to assessing application-specific explainability needs for the design of LCS models. For this, we propose an application-independent template of seven questions. We demonstrate the approach's use in an interview-based case study for a manufacturing scenario. We find that the answers received do yield useful insights for a well-designed LCS model and requirements for stakeholders to engage actively with an intelligent agent.

To fill the increasing demand for explanations of decisions made by automated prediction systems, machine learning (ML) techniques that produce inherently transparent models are directly suited. Learning Classifier Systems (LCSs), a family of rule-based learners, produce transparent models by design. However, the usefulness of such models, both for predictions and analyses, heavily depends on the placement and selection of rules (combined constituting the ML task of model selection). In this paper, we investigate a variety of techniques to efficiently place good rules within the search space based on their local prediction errors as well as their generality. This investigation is done within a specific LCS, named SupRB, where the placement of rules and the selection of good subsets of rules are strictly separated in contrast to other LCSs where these tasks sometimes blend. We compare a Random Search, (1,λ)-ES and three Novelty Search variants. We find that there is a definitive need to guide the search based on some sensible criteria, i.e. error and generality, rather than just placing rules randomly and selecting better performing ones but also find that Novelty Search variants do not beat the easier to understand (1,λ)-ES.

Abstract Metaheuristics provide the means to approximately solve complex optimisation problems when exact optimisers cannot be utilised. This led to an explosion in the number of novel metaheuristics, most of them metaphor-based, using nature as a source of inspiration. Thus, keeping track of their capabilities and innovative components is an increasingly difficult task. This can be resolved by an exhaustive classification system. Trying to classify metaheuristics is common in research, but no consensus on a classification system and the necessary criteria has been established so far. Furthermore, a proposed classification system can not be deemed complete if inherently different metaheuristics are assigned to the same class by the system. In this paper we provide the basis for a new comprehensive classification system for metaheuristics. We first summarise and discuss previous classification attempts and the utilised criteria. Then we present a multi-level architecture and suitable criteria for the task of classifying metaheuristics. A classification system of this kind can solve three main problems when applied to metaheuristics: organise the huge set of existing metaheuristics, clarify the innovation in novel metaheuristics and identify metaheuristics suitable to solve specific optimisation tasks.

Planning the right grasp pose and motion into it has been a problem in the robotic community for more than 20 years.This paper presents a model-based approach for a Pick action of a robot that increases the automation of FDM based additive manufacturing by removing a produced object from the build plate.We treat grasp pose planning, motion planning and simulation-based verification as separate components to allow a high exchangeability.When testing a variety of different object geometries, feasible grasps and motions were obtained for all objects.We also found that the computation time is highly dependent on the random seed, leading us to employ a system of budgeted runs for which we report the estimated success probability and expected running time.Within the budget, some objects never found feasible picks.Thus, we rotated these objects by 90 • which lead to a substantial improvement in success probabilities.

Commissioning of machines takes up a considerable share of time and money of the total cost of developing a machine.Our project aims at developing an approach to decrease the time needed to commission machines by automating parameter optimisation with the help of formalised expert knowledge.The approach will be developed on the Fused Deposition Modelling (FDM) process, which is an additive manufacturing technique.We pay particular attention to keeping the approach sufficiently abstract to be applied to machines from other domains to benefit its industrial application.

Robot Gardens are an augmented reality concept allowing a human user to design a biohybrid, plant-robot system. Plants growing from deliberately placed seeds are directed by robotic units that the user can position, configure and activate. For example, the robotic units may serve as physical shields or frames but they may also guide the plants' growth through emission of light. The biohybrid system evolves over time to redefine architectural spaces. This gives rise to the particular challenge of designing a biohybrid system before its actual implementation and potentially long before its developmental processes unfold. Here, an augmented reality interface featuring according simulation models of plants and robotic units allows one to explore the design space a priori. In this work, we present our first functional augmented reality prototype to design biohybrid systems. We provide details about its workings and elaborate on first empirical studies on its usability.