Ilias Gerostathopoulos

Erik M. Fredericks, Ilias Gerostathopoulos, Christian Krupitzer et al.

The large number of possible configurations of modern software-based systems, combined with the large number of possible environmental situations of such systems, prohibits enumerating all adaptation options at design time and necessitates planning at run time to dynamically identify an appropriate configuration for a situation. While numerous planning techniques exist, they typically assume a detailed state-based model of the system and that the situations that warrant adaptations are known. Both of these assumptions can be violated in complex, real-world systems. As a result, adaptation planning must rely on simple models that capture what can be changed (input parameters) and observed in the system and environment (output and context parameters). We therefore propose planning as optimization: the use of optimization strategies to discover optimal system configurations at runtime for each distinct situation that is also dynamically identified at runtime. We apply our approach to CrowdNav, an open-source traffic routing system with the characteristics of a real-world system. We identify situations via clustering and conduct an empirical study that compares Bayesian optimization and two types of evolutionary optimization (NSGA-II and novelty search) in CrowdNav.

Zakkarija Micallef, Keerthiga Rajenthiram, Ilias Gerostathopoulos

Machine Learning Operations (MLOps) has become increasingly critical as more organisations move ML models into production. However, the growing landscape of MLOps solutions has introduced complexity for practitioners trying to select appropriate tools. To investigate how and why these tools are adopted in practice, this paper conducts a systematic review of the academic literature focused on MLOps tools. We map tools to MLOps lifecycle components to reveal their function, scope, and the challenges they are designed to address. We identify usage trends and synthesise reported benefits and limitations. The most commonly used components, according to the findings, are orchestration frameworks, data versioning, experiment tracking, and managed cloud platforms. No single tool covers the entire lifecycle, so researchers often combine multiple tools to build complete pipelines. This highlights the importance of interoperability across MLOps tools in real-world MLOps pipelines.

Shreyas Kumar Parida, Ilias Gerostathopoulos, Justus Bogner

Machine learning (ML) models are often integrated into ML-enabled systems to provide software functionality that would otherwise be impossible. This integration requires the selection of an appropriate ML model export format, for which many options are available. These formats are crucial for ensuring a seamless integration, and choosing a suboptimal one can negatively impact system development. However, little evidence is available to guide practitioners during the export format selection. We therefore evaluated various model export formats regarding their impact on the development of ML-enabled systems from an integration perspective. Based on the results of a preliminary questionnaire survey (n=17), we designed an extensive embedded case study with two ML-enabled systems in three versions with different technologies. We then analyzed the effect of five popular export formats, namely ONNX, Pickle, TensorFlow's SavedModel, PyTorch's TorchScript, and Joblib. In total, we studied 30 units of analysis (2 systems x 3 tech stacks x 5 formats) and collected data via structured field notes. The holistic qualitative analysis of the results indicated that ONNX offered the most efficient integration and portability across most cases. SavedModel and TorchScript were very convenient to use in Python-based systems, but otherwise required workarounds (TorchScript more than SavedModel). SavedModel also allowed the easy incorporation of preprocessing logic into a single file, which made it scalable for complex deep learning use cases. Pickle and Joblib were the most challenging to integrate, even in Python-based systems. Regarding technical support, all model export formats had strong technical documentation and strong community support across platforms such as Stack Overflow and Reddit. Practitioners can use our findings to inform the selection of ML export formats suited to their context.

Joran Leest, Claudia Raibulet, Ilias Gerostathopoulos et al.

We propose Expert Monitoring, an approach that leverages domain expertise to enhance the detection and mitigation of concept drift in machine learning (ML) models. Our approach supports practitioners by consolidating domain expertise related to concept drift-inducing events, making this expertise accessible to on-call personnel, and enabling automatic adaptability with expert oversight.

Tomáš Bureš, Ilias Gerostathopoulos, Petr Hnětynka et al.

Smart system applications (SSAs) built on top of cyber-physical and socio-technical systems are increasingly composed of components that can work both autonomously and by cooperating with each other. Cooperating robots, fleets of cars and fleets of drones, emergency coordination systems are examples of SSAs. One approach to enable cooperation of SSAs is to form dynamic cooperation groups-ensembles-between components at runtime. Ensembles can be formed based on predefined rules that determine which components should be part of an ensemble based on their current state and the state of the environment (e.g., "group together 3 robots that are closer to the obstacle, their battery is sufficient and they would not be better used in another ensemble"). This is a computationally hard problem since all components are potential members of all possible ensembles at runtime. In our experience working with ensembles in several case studies the past years, using constraint programming to decide which ensembles should be formed does not scale for more than a limited number of components and ensembles. Also, the strict formulation in terms of hard/soft constraints does not easily permit for runtime self-adaptation via learning. This poses a serious limitation to the use of ensembles in large-scale and partially uncertain SSAs. To tackle this problem, in this paper we propose to recast the ensemble formation problem as a classification problem and use machine learning to efficiently form ensembles at scale.

Ilias Gerostathopoulos, Thomas Vogel, Danny Weyns et al.

With the increase of research in self-adaptive systems, there is a need to better understand the way research contributions are evaluated. Such insights will support researchers to better compare new findings when developing new knowledge for the community. However, so far there is no clear overview of how evaluations are performed in self-adaptive systems. To address this gap, we conduct a mapping study. The study focuses on experimental evaluations published in the last decade at the prime venue of research in software engineering for self-adaptive systems -- the International Symposium on Software Engineering for Adaptive and Self-Managing Systems (SEAMS). Results point out that specifics of self-adaptive systems require special attention in the experimental process, including the distinction of the managing system (i.e., the target of evaluation) and the managed system, the presence of uncertainties that affect the system behavior and hence need to be taken into account in data analysis, and the potential of managed systems to be reused across experiments, beyond replications. To conclude, we offer a set of suggestions derived from our study that can be used as input to enhance future experiments in self-adaptive systems.

Justus Bogner, Roberto Verdecchia, Ilias Gerostathopoulos

Background: With the rising popularity of Artificial Intelligence (AI), there is a growing need to build large and complex AI-based systems in a cost-effective and manageable way. Like with traditional software, Technical Debt (TD) will emerge naturally over time in these systems, therefore leading to challenges and risks if not managed appropriately. The influence of data science and the stochastic nature of AI-based systems may also lead to new types of TD or antipatterns, which are not yet fully understood by researchers and practitioners. Objective: The goal of our study is to provide a clear overview and characterization of the types of TD (both established and new ones) that appear in AI-based systems, as well as the antipatterns and related solutions that have been proposed. Method: Following the process of a systematic mapping study, 21 primary studies are identified and analyzed. Results: Our results show that (i) established TD types, variations of them, and four new TD types (data, model, configuration, and ethics debt) are present in AI-based systems, (ii) 72 antipatterns are discussed in the literature, the majority related to data and model deficiencies, and (iii) 46 solutions have been proposed, either to address specific TD types, antipatterns, or TD in general. Conclusions: Our results can support AI professionals with reasoning about and communicating aspects of TD present in their systems. Additionally, they can serve as a foundation for future research to further our understanding of TD in AI-based systems.

Brionna Davis, Grace Jennings, Taylor Pothast et al.

New mobility concepts are at the forefront of research and innovation in smart cities. The introduction of connected and autonomous vehicles enables new possibilities in vehicle routing. Specifically, knowing the origin and destination of each agent in the network can allow for real-time routing of the vehicles to optimize network performance. However, this relies on individual vehicles being "altruistic" i.e., being willing to accept an alternative non-preferred route in order to achieve a network-level performance goal. In this work, we conduct a study to compare different levels of agent altruism and the resulting effect on the network-level traffic performance. Specifically, this study compares the effects of different underlying urban structures on the overall network performance, and investigates which characteristics of the network make it possible to realize routing improvements using a decentralized optimization router. The main finding is that, with increased vehicle altruism, it is possible to balance traffic flow among the links of the network. We show evidence that the decentralized optimization router is more effective with networks of high load while we study the influence of cities characteristics, in particular: networks with a higher number of nodes (intersections) or edges (roads) per unit area allow for more possible alternate routes, and thus higher potential to improve network performance.

Nikos Diamantopoulos, Jeffrey Wong, David Issa Mattos et al.

Netflix is an internet entertainment service that routinely employs experimentation to guide strategy around product innovations. As Netflix grew, it had the opportunity to explore increasingly specialized improvements to its service, which generated demand for deeper analyses supported by richer metrics and powered by more diverse statistical methodologies. To facilitate this, and more fully harness the skill sets of both engineering and data science, Netflix engineers created a science-centric experimentation platform that leverages the expertise of data scientists from a wide range of backgrounds by allowing them to make direct code contributions in the languages used by scientists (Python and R). Moreover, the same code that runs in production is able to be run locally, making it straightforward to explore and graduate both metrics and causal inference methodologies directly into production services. In this paper, we utilize a case-study research method to provide two main contributions. Firstly, we report on the architecture of this platform, with a special emphasis on its novel aspects: how it supports science-centric end-to-end workflows without compromising engineering requirements. Secondly, we describe its approach to causal inference, which leverages the potential outcomes conceptual framework to provide a unified abstraction layer for arbitrary statistical models and methodologies.