Continuous Management of Machine Learning-Based Application Behavior

被引:0
作者
Anisetti, Marco [1 ]
Ardagna, Claudio A. [1 ]
Bena, Nicola [1 ]
Damiani, Ernesto [1 ,2 ]
Panero, Paolo G. [1 ]
机构
[1] Univ Milan, Dept Comp Sci, I-20133 Milan, Italy
[2] Khalifa Univ, Comp Sci Dept, C2PS, POB 127788, Abu Dhabi, U Arab Emirates
关键词
Data models; Context modeling; Degradation; Computational modeling; Windows; Monitoring; Accuracy; Measurement; Classification algorithms; Training; Assurance; machine learning; multi-armed bandit; non-functional properties; DYNAMIC CLASSIFIER SELECTION;
D O I
10.1109/TSC.2024.3486226
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Modern applications are increasingly driven by Machine Learning (ML) models whose non-deterministic behavior is affecting the entire application life cycle from design to operation. The pervasive adoption of ML is urgently calling for approaches that guarantee a stable non-functional behavior of ML-based applications over time and across model changes. To this aim, non-functional properties of ML models, such as privacy, confidentiality, fairness, and explainability, must be monitored, verified, and maintained. Existing approaches mostly focus on i) implementing solutions for classifier selection according to the functional behavior of ML models, ii) finding new algorithmic solutions, such as continuous re-training. In this paper, we propose a multi-model approach that aims to guarantee a stable non-functional behavior of ML-based applications. An architectural and methodological approach is provided to compare multiple ML models showing similar non-functional properties and select the model supporting stable non-functional behavior over time according to (dynamic and unpredictable) contextual changes. Our approach goes beyond the state of the art by providing a solution that continuously guarantees a stable non-functional behavior of ML-based applications, is ML algorithm-agnostic, and is driven by non-functional properties assessed on the ML models themselves. It consists of a two-step process working during application operation, where model assessment verifies non-functional properties of ML models trained and selected at development time, and model substitution guarantees continuous and stable support of non-functional properties. We experimentally evaluate our solution in a real-world scenario focusing on non-functional property fairness.
引用
收藏
页码:112 / 125
页数:14
相关论文
共 50 条
  • [21] Machine Learning-Based Classification of Academic Performance via Imaging Sensors
    Lin, Yongzheng
    Liu, Hong
    Chen, Zhenxiang
    Ma, Kun
    [J]. IEEE SENSORS JOURNAL, 2021, 21 (22) : 24952 - 24958
  • [22] Adversarial Attack Mitigation Strategy for Machine Learning-Based Network Attack Detection Model in Power System
    Huang, Rong
    Li, Yuancheng
    [J]. IEEE TRANSACTIONS ON SMART GRID, 2023, 14 (03) : 2367 - 2376
  • [23] Simple Yet Powerful: Machine Learning-Based IoT Intrusion System With Smart Preprocessing and Feature Generation Rivals Deep Learning
    Kivanc Eren, Kazim
    Kucuk, Kerem
    Ozyurt, Fatih
    Alhazmi, Omar H.
    [J]. IEEE ACCESS, 2025, 13 : 41435 - 41455
  • [24] Data Management for Machine Learning: A Survey
    Chai, Chengliang
    Wang, Jiayi
    Luo, Yuyu
    Niu, Zeping
    Li, Guoliang
    [J]. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2023, 35 (05) : 4646 - 4667
  • [25] Continuous Defect Prediction in CI/CD Pipelines: A Machine Learning-Based Framework
    Giorgio, Lazzarinetti
    Nicola, Massarenti
    Fabio, Sgro
    Andrea, Salafia
    [J]. AIXIA 2021 - ADVANCES IN ARTIFICIAL INTELLIGENCE, 2022, 13196 : 591 - 606
  • [26] Efficient Learning Strategies for Machine Learning-Based Characterization of Aging-Aware Cell Libraries
    Klemme, Florian
    Amrouch, Hussam
    [J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS, 2022, 69 (12) : 5233 - 5246
  • [27] Machine learning-based methods for TTF estimation with application to APU prognostics
    Yang, Chunsheng
    Letourneau, Sylvain
    Liu, Jie
    Cheng, Qiangqiang
    Yang, Yubin
    [J]. APPLIED INTELLIGENCE, 2017, 46 (01) : 227 - 239
  • [28] Development and application of a machine learning-based antenatal depression prediction model
    Hu, Chunfei
    Lin, Hongmei
    Xu, Yupin
    Fu, Xukun
    Qiu, Xiaojing
    Hu, Siqian
    Jin, Tong
    Xu, Hualin
    Luo, Qiong
    [J]. JOURNAL OF AFFECTIVE DISORDERS, 2025, 375 : 137 - 147
  • [29] Machine learning-based methods for TTF estimation with application to APU prognostics
    Chunsheng Yang
    Sylvain Letourneau
    Jie Liu
    Qiangqiang Cheng
    Yubin Yang
    [J]. Applied Intelligence, 2017, 46 : 227 - 239
  • [30] Development and application of machine learning-based prediction model for distillation column
    Kwon, Hyukwon
    Oh, Kwang Cheol
    Choi, Yeongryeol
    Chung, Yongchul G.
    Kim, Junghwan
    [J]. INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS, 2021, 36 (05) : 1970 - 1997