mlsanityData to 7 Oct 2026

Tunability: Importance of Hyperparameters of Machine Learning Algorithms

arXiv 1802.09596 ↗First posted 26 Feb 2018stat.MLcited by 98

Not on the map. Why?

Abstract (arXiv)

Modern supervised machine learning algorithms involve hyperparameters that have to be set before running them. Options for setting hyperparameters are default values from the software package, manual configuration by the user or configuring them for optimal predictive performance by a tuning procedure. The goal of this paper is two-fold. Firstly, we formalize the problem of tuning from a statistical point of view, define data-based defaults and suggest general measures quantifying the tunability of hyperparameters of algorithms. Secondly, we conduct a large-scale benchmarking study based on 38 datasets from the OpenML platform and six common machine learning algorithms. We apply our measures to assess the tunability of their parameters. Our results yield default values for hyperparameters and enable users to decide whether it is worth conducting a possibly time consuming tuning strategy, to focus on the most important hyperparameters and to chose adequate hyperparameter spaces for tuning.

Key facts. Cited by 98 papers in the map (data through 7 Oct 2026). First posted on arXiv 26 Feb 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 17; 2022: 14; 2023: 13; 2024: 19; 2025: 16; 2026: 19.

98 papers.

Sort:

  1. 2021Hyperparameter Optimization: Foundations, Algorithms, Best Practices and Open Challengescited by 143
  2. 2021The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanationscited by 86
  3. 2022AMLB: an AutoML Benchmarkcited by 55
  4. 2022ADATIME: A Benchmarking Suite for Domain Adaptation on Time Series Datacited by 45
  5. 2022Automated Dynamic Algorithm Configurationcited by 37
  6. 2021Explaining Hyperparameter Optimization via Partial Dependence Plotscited by 35
  7. 2021How to Train Your MAML to Excel in Few-Shot Classificationcited by 22
  8. 2025Optuna vs Code Llama: Are LLMs a New Paradigm for Hyperparameter Tuning?cited by 21
  9. 2023Can Fairness be Automated? Guidelines and Opportunities for Fairness-aware AutoMLcited by 16
  10. 2024Position: A Call to Action for a Human-Centered AutoML Paradigmcited by 14
  11. 2021Trees, Forests, Chickens, and Eggs: When and Why to Prune Trees in a Random Forestcited by 13
  12. 2021Automated Evolutionary Approach for the Design of Composite Machine Learning Pipelinescited by 13
38 more
  1. 2023REFORMS: Reporting Standards for Machine Learning Based Sciencecited by 13
  2. 2022HPO X ELA: Investigating Hyperparameter Optimization Landscapes by Means of Exploratory Landscape Analysiscited by 9
  3. 2024Hyperparameter Importance Analysis for Multi-Objective AutoMLcited by 9
  4. 2024Scikit-fingerprints: easy and efficient computation of molecular fingerprints in Pythoncited by 9
  5. 2022KGTuner: Efficient Hyper-parameter Search for Knowledge Graph Learningcited by 7
  6. 2023Strengthening structural baselines for graph classification using Local Topological Profilecited by 7
  7. 2024Reshuffling Resampling Splits Can Improve Generalization of Hyperparameter Optimizationcited by 7
  8. 2024Molecular Topological Profile (MOLTOP) -- Simple and Strong Baseline for Molecular Graph Classificationcited by 7
  9. 2021When are Deep Networks really better than Decision Forests at small sample sizes, and how?cited by 6
  10. 2022Ultra-marginal Feature Importance: Learning from Data with Causal Guaranteescited by 6
  11. 2023Multi-Objective Optimization of Performance and Interpretability of Tabular Supervised Machine Learning Modelscited by 6
  12. 2021Using AntiPatterns to avoid MLOps Mistakescited by 5
  13. 2021Test for non-negligible adverse shiftscited by 5
  14. 2024Maintainability Challenges in ML: A Systematic Literature Reviewcited by 5
  15. 2025Molecular Fingerprints Are Strong Models for Peptide Function Predictioncited by 5
  16. 2025How Should We Meta-Learn Reinforcement Learning Algorithms?cited by 5
  17. 2021Meta-Learning for Symbolic Hyperparameter Defaultscited by 4
  18. 2022Hyperparameter Importance for Machine Learning Algorithmscited by 4
  19. 2022What Makes Forest-Based Heterogeneous Treatment Effect Estimators Work?cited by 4
  20. 2024Machine Learning Approaches for Mental Illness Detection on Social Media: A Systematic Review of Biases and Methodological Challengescited by 4
  21. 2025HyperSHAP: Shapley Values and Interactions for Explaining Hyperparameter Optimizationcited by 4
  22. 2025The Role of Hyperparameters in Predictive Multiplicitycited by 4
  23. 2025Hyperparameter Optimization via Interacting with Probabilistic Circuitscited by 4
  24. 2025Overtuning in Hyperparameter Optimizationcited by 4
  25. 2024TERC: A Transfer Entropy Redundancy Criterion for State Variable Selection in Reinforcement Learningcited by 3
  26. 2024TensorSocket: Shared Data Loading for Deep Learning Trainingcited by 3
  27. 2021Exploring Opportunistic Meta-knowledge to Reduce Search Spaces for Automated Machine Learningcited by 2
  28. 2021To tune or not to tune? An Approach for Recommending Important Hyperparameterscited by 2
  29. 2022Consolidated learning -- a domain-specific model-free optimization strategy with examples for XGBoost and MIMIC-IVcited by 2
  30. 2022Improving Accuracy of Interpretability Measures in Hyperparameter Optimization via Bayesian Algorithm Executioncited by 2
  31. 2023Controlling Federated Learning for Covertnesscited by 2
  32. 2024Enhancing literature review with LLM and NLP methods. Algorithmic trading casecited by 2
  33. 2025Foundation for unbiased cross-validation of spatio-temporal models for species distribution modelingcited by 2
  34. 2025Evaluating machine learning models for predicting pesticide toxicity to honey beescited by 2
  35. 2025Unsupervised Machine Learning for Scientific Discovery: Workflow and Best Practicescited by 2
  36. 2022High Per Parameter: A Large-Scale Study of Hyperparameter Tuning for Machine Learning Algorithmscited by 1
  37. 2023Towards Personalized Preprocessing Pipeline Searchcited by 1
  38. 2023Comparing statistical and machine learning methods for time series forecasting in data-driven logistics -- A simulation studycited by 1

50 most cited of 98 All 98 (JSON)