mlsanityData to 14 Oct 2026

Test & Evaluation Best Practices for Machine Learning-Enabled Systems

arXiv 2310.06800 ↗First posted 10 Oct 2023cs.SE cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Machine learning (ML) - based software systems are rapidly gaining adoption across various domains, making it increasingly essential to ensure they perform as intended. This report presents best practices for the Test and Evaluation (T&E) of ML-enabled software systems across its lifecycle. We categorize the lifecycle of ML-enabled software systems into three stages: component, integration and deployment, and post-deployment. At the component level, the primary objective is to test and evaluate the ML model as a standalone component. Next, in the integration and deployment stage, the goal is to evaluate an integrated ML-enabled system consisting of both ML and non-ML components. Finally, once the ML-enabled software system is deployed and operationalized, the T&E objective is to ensure the system performs as intended. Maintenance activities for ML-enabled software systems span the lifecycle and involve maintaining various assets of ML-enabled software systems. Given its unique characteristics, the T&E of ML-enabled software systems is challenging. While significant research has been reported on T&E at the component level, limited work is reported on T&E in the remaining two stages. Furthermore, in many cases, there is a lack of systematic T&E strategies throughout the ML-enabled system's lifecycle. This leads practitioners to resort to ad-hoc T&E practices, which can undermine user confidence in the reliability of ML-enabled software systems. New systematic testing approaches, adequacy measurements, and metrics are required to address the T&E challenges across all stages of the ML-enabled system lifecycle.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 45 of its 105 references that are in the map. First posted on arXiv 10 Oct 2023.

Builds on

45 of 105 references.

References · 45

  1. 2013Intriguing properties of neural networkscited by 5,155
  2. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 661
  3. 2020Learning under Concept Drift: A Reviewcited by 477
  4. 2017DeepXplore: Automated Whitebox Testing of Deep Learning Systemscited by 311
  5. 2017DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Carscited by 251
  6. 2019Machine Learning Testing: Survey, Landscapes and Horizonscited by 218
  7. 2018DeepGauge: Multi-Granularity Testing Criteria for Deep Learning Systemscited by 147
  8. 2018Guiding Deep Learning System Testing using Surprise Adequacycited by 124
  9. 2015Characterizing Concept Driftcited by 94
  10. 2022From Concept Drift to Model Degradation: An Overview on Performance-Aware Drift Detectorscited by 83
  11. 2018TensorFuzz: Debugging Neural Networks with Coverage-Guided Fuzzingcited by 68
  12. 2018DLFuzz: Differential Fuzzing Testing of Deep Learning Systemscited by 65
33 more
  1. 2021Collaboration Challenges in Building ML-Enabled Systems: Communication, Documentation, Engineering, and Processcited by 56
  2. 2015MLlib: Machine Learning in Apache Sparkcited by 55
  3. 2018DeepBillboard: Systematic Physical-World Testing of Autonomous Driving Systemscited by 50
  4. 2022Evaluation Gaps in Machine Learning Practicecited by 49
  5. 2019An Empirical Study towards Characterizing Deep Learning Development and Deployment across Different Frameworks and Platformscited by 35
  6. 2020Repairing Deep Neural Networks: Fix Patterns and Challengescited by 33
  7. 2020A Software Engineering Perspective on Engineering Machine Learning Systems: State of the Art and Challengescited by 33
  8. 2018Identifying Implementation Bugs in Machine Learning based Image Classifiers using Metamorphic Testingcited by 31
  9. 2018Simulation-based Adversarial Test Generation for Autonomous Vehicles with Machine Learning Componentscited by 29
  10. 2017Failing to Learn: Autonomously Identifying Perception Failures for Self-driving Carscited by 28
  11. 2022Free Lunch for Testing: Fuzzing Deep-Learning Libraries from Open Sourcecited by 28
  12. 2020More Than Privacy: Applying Differential Privacy in Key Areas of Artificial Intelligencecited by 23
  13. 2021DocTer: Documentation Guided Fuzzing for Testing Deep Learning API Functionscited by 22
  14. 2023A Meta-Summary of Challenges in Building Products with ML Components -- Collecting Experiences from 4758+ Practitionerscited by 22
  15. 2021An Empirical Study on Deployment Faults of Deep Learning Based Mobile Applicationscited by 20
  16. 2020Detection of data drift and outliers affecting machine learning model performance over timecited by 19
  17. 2023Fuzzing Automatic Differentiation in Deep-Learning Librariescited by 17
  18. 2022Fuzzing Deep-Learning Libraries via Automated Relational API Inferencecited by 16
  19. 2021Automatically detecting data drift in machine learning classifierscited by 15
  20. 2023Challenges and Practices of Deep Learning Model Reengineering: A Case Study on Computer Visioncited by 11
  21. 2021FreaAI: Automated extraction of data slices to test machine learning modelscited by 10
  22. 2018METTLE: a METamorphic testing approach to assessing and validating unsupervised machine LEarning systemscited by 8
  23. 2022An Empirical Study of Challenges in Converting Deep Learning Modelscited by 8
  24. 2022Are Concept Drift Detectors Reliable Alarming Systems? -- A Comparative Studycited by 7
  25. 2022Systematic Training and Testing for Machine Learning Using Combinatorial Interaction Testingcited by 6
  26. 2023AI Maintenance: A Robustness Perspectivecited by 6
  27. 2019Software Engineering Practice in the Development of Deep Learning Applicationscited by 5
  28. 2020Smoke Testing for Machine Learning: Simple Tests to Discover Severe Defectscited by 5
  29. 2018DeepCruiser: Automated Guided Testing for Stateful Deep Learning Systemscited by 4
  30. 2022Differential testing for machine learning: an analysis for classification algorithms beyond deep learningcited by 3
  31. 2023Active Learning with Combinatorial Coveragecited by 3
  32. 2023MLOps Spanning Whole Machine Learning Life Cycle: A Surveycited by 3
  33. 2022Testing Deep Learning Models: A First Comparative Study of Multiple Testing Techniquescited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 2.

4 papers.

  1. 2025An Empirical Study of Testing Practices in Open Source AI Agent Frameworks and Agentic Applicationscited by 15
  2. 2026Evaluation Cards: An Interpretive Layer for AI Evaluation Reportingcited by 4
  3. 2025A Guide to Failure in Machine Learning: Reliability and Robustness from Foundations to Practicecited by 0
  4. 2026Quality Model for Machine Learning Componentscited by 0