Fuzzy evolutionary systems: first steps towards supporting model lifecycle

Abstract only given. The typical lifecycle of a knowledge-based model starts from its development, testing, optimization, and deployment, and continues with its maintenance phase. The latter consists of monitoring the model's performance, editing its knowledge base to prevent obsolescence, and updating the model when required. Quite often, however, models are handcrafted, i.e., a large amount of manual intervention is used in the earlier phase of their lifecycle. This leaves the maintenance phase as an after-thought, often requiring a similar level of manual efforts. We propose a process, based on evolutionary algorithms, that follows the model throughout its entire lifecycle. For deployment, it generates a collection of competing models, evaluates their performance in light of the currently available data, refines the best models using evolutionary search, and after a finite number of iterations, generates the best-found model. This process is repeated periodically to automatically produce new updated versions of the model. We chose an asset selection problem to illustrate this methodology. Given a fleet of industrial vehicles (diesel electric locomotives), we want to select the best subset (of fixed or variable size) for mission-critical utilization. To this end, we predict the remaining life for each unit in the fleet. We then sort the fleet using this prediction and select the highest ranked units. The model chosen to perform this prediction/selection task is a fuzzy instance-based model. Unlike functional approximators that require an off-line supervised learning stage to create a hyper-surface in the cross-space state-output, instance-based models (IBMs) only require accessing the data in a local neighbourhood of the new point defined by the query. IBMs rely on a collection of previously experienced data stored in their raw representation. Unlike case-based reasoning (CBR), they do not need to be refined, abstracted and organized as cases. Like CBR, IBMs represent an analogical approach to reasoning since they rely on previous instances of similar problems and use them to create an ensemble of local models. Hence the definition of similarity plays a critical role in their performance. Typically, similarity will be a dynamic concept and will change over the use of the IBMs. Therefore, it is important to apply learning methodologies to define and adapt it. Furthermore, the concept of similarity is not crisply defined, creating the need to allow for some degree of vagueness in its evaluation. Hence, we propose the use of fuzzy IBMs (F-IBMs). We address the issue of similarity by evolving the design of a similarity function in conjunction with the design of the attribute space in which the similarity is to be evaluated. Specifically, we use four steps: (1) Retrieval of similar instances from the database (DB); (2) Evaluation of similarity measures between the probe and the retrieved instances; (3) Creation of local models using the most similar instances (weighted by their similarity measures); (4) Aggregation of outputs of local models to probe. Within the example of asset selection, we show the accuracy of the evolved F-IBMs, their robustness to information loss, and the benefit of their automated updating process to avoid performance loss. Finally, we advocate the use of evolutionary search intertwined with local search to further improve model life cycle.