mlsanityData to 14 Oct 2026

Automated Dynamic Algorithm Configuration

arXiv 2205.13881 ↗First posted 27 May 2022cs.AI cs.LG cs.NEcited by 39

Not on the map. Why?

Abstract (arXiv)

The performance of an algorithm often critically depends on its parameter configuration. While a variety of automated algorithm configuration methods have been proposed to relieve users from the tedious and error-prone task of manually tuning parameters, there is still a lot of untapped potential as the learned configuration is static, i.e., parameter settings remain fixed throughout the run. However, it has been shown that some algorithm parameters are best adjusted dynamically during execution, e.g., to adapt to the current part of the optimization landscape. Thus far, this is most commonly achieved through hand-crafted heuristics. A promising recent alternative is to automatically learn such dynamic parameter adaptation policies from data. In this article, we give the first comprehensive account of this new field of automated dynamic algorithm configuration (DAC), present a series of recent advances, and provide a solid foundation for future research in this field. Specifically, we (i) situate DAC in the broader historical context of AI research; (ii) formalize DAC as a computational problem; (iii) identify the methods used in prior-art to tackle this problem; (iv) conduct empirical case studies for using DAC in evolutionary optimization, AI planning, and machine learning.

Key facts. Cited by 39 papers in the map (data through 14 Oct 2026). Builds on 52 of its 175 references that are in the map. First posted on arXiv 27 May 2022.

Builds on

52 of 175 references.

References · 52

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  3. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  4. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,745
  5. 2017Deep Reinforcement Learning that Matterscited by 754
  6. 2016Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimizationcited by 699
  7. 2018Attention, Learn to Solve Routing Problems!cited by 656
  8. 2015Cyclical Learning Rates for Training Neural Networkscited by 629
  9. 2016Learning to learn by gradient descent by gradient descentcited by 620
  10. 2020Learning Quadrupedal Locomotion over Challenging Terraincited by 604
  11. 2017Evolution Strategies as a Scalable Alternative to Reinforcement Learningcited by 574
  12. 2018BOHB: Robust and Efficient Hyperparameter Optimization at Scalecited by 470
38 more
  1. 2015Gradient-based Hyperparameter Optimization through Reversible Learningcited by 360
  2. 2021SMAC3: A Versatile Bayesian Optimization Package for Hyperparameter Optimizationcited by 230
  3. 2015Contextual Markov Decision Processescited by 220
  4. 2021Learning to Optimize: A Primer and A Benchmarkcited by 208
  5. 2016Learning to Optimizecited by 195
  6. 2013Bayesian Optimization in a Billion Dimensions via Random Embeddingscited by 182
  7. 2017Online Learning Rate Adaptation with Hypergradient Descentcited by 118
  8. 2018Automated Algorithm Selection: Survey and Perspectivescited by 114
  9. 2018Tunability: Importance of Hyperparameters of Machine Learning Algorithmscited by 99
  10. 2018Understanding and correcting pathologies in the training of learned optimizerscited by 97
  11. 2014ParamILS: An Automatic Algorithm Configuration Frameworkcited by 89
  12. 2017Neural Optimizer Search with Reinforcement Learningcited by 82
  13. 2022πBO: Augmenting Acquisition Functions with User Beliefs for Bayesian Optimizationcited by 70
  14. 2017Hyperparameter Importance Across Datasetscited by 63
  15. 2020Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselvescited by 58
  16. 2017Learning Gradient Descent: Better Generalization and Longer Horizonscited by 50
  17. 2020Self-Paced Deep Reinforcement Learningcited by 50
  18. 2019Deep Reinforcement Learning Based Parameter Control in Differential Evolutioncited by 38
  19. 2015ASlib: A Benchmark Library for Algorithm Selectioncited by 37
  20. 2021Accelerating Quadratic Optimization with Reinforcement Learningcited by 30
  21. 2019Learning an Adaptive Learning Rate Schedulecited by 29
  22. 2018Back to Basics: Benchmarking Canonical Evolution Strategies for Playing Ataricited by 28
  23. 2018IOHprofiler: A Benchmarking and Profiling Tool for Iterative Optimization Heuristicscited by 28
  24. 2021DACBench: A Benchmark Library for Dynamic Algorithm Configurationcited by 28
  25. 2021Deep Reinforcement Learning Versus Evolution Strategies: A Comparative Surveycited by 28
  26. 2021Self-Paced Context Evaluation for Contextual Reinforcement Learningcited by 25
  27. 2021TempoRL: Learning When to Actcited by 25
  28. 2021A Generalizable Approach to Learning Optimizerscited by 23
  29. 2020Learning Heuristic Selection with Dynamic Algorithm Configurationcited by 21
  30. 2017Reinforcement Learning for Learning Rate Controlcited by 16
  31. 2019Online Selection of CMA-ES Variantscited by 11
  32. 2022Theory-inspired Parameter Control Benchmarks for Dynamic Algorithm Configurationcited by 11
  33. 2018Theory of Parameter Control for Discrete Black-Box Optimization: Provable Performance Gains Through Dynamic Parameter Choicescited by 9
  34. 2014Solver Scheduling via Answer Set Programmingcited by 7
  35. 2016Using Deep Q-Learning to Control Optimization Hyperparameterscited by 4
  36. 2019Learning Neural Search Policies for Classical Planningcited by 4
  37. 2019On the Impact of the Cutoff Time on the Performance of Algorithm Configuratorscited by 3
  38. 2018Learning Classical Planning Strategies with Policy Gradientcited by 2

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 2; 2023: 9; 2024: 9; 2025: 11; 2026: 8.

39 papers.

Sort:

By area:

  1. 2023Hyperparameters in Reinforcement Learning and How To Tune Themcited by 59
  2. 2022Multi-agent Dynamic Algorithm Configurationcited by 36
  3. 2023AutoML in the Age of Large Language Models: Current Challenges, Future Opportunities and Riskscited by 35
  4. 2024Toward Automated Algorithm Design: A Survey and Practical Guide to Meta-Black-Box-Optimizationcited by 24
  5. 2024Reinforced In-Context Black-Box Optimizationcited by 14
  6. 2024Position: A Call to Action for a Human-Centered AutoML Paradigmcited by 14
  7. 2024Artificial Intelligence for Operations Research: Revolutionizing the Operations Research Processcited by 11
  8. 2023To Switch or not to Switch: Predicting the Benefit of Switching between Algorithms based on Trajectory Featurescited by 9
  9. 2023AutoRL Hyperparameter Landscapescited by 8
  10. 2024Learning to Stop Cut Generation for Efficient Mixed-Integer Linear Programmingcited by 8
  11. 2022PI is back! Switching Acquisition Functions in Bayesian Optimizationcited by 6
  12. 2023Using Automated Algorithm Configuration for Parameter Controlcited by 5
27 more
  1. 2023Self-Adjusting Weighted Expected Improvement for Bayesian Optimizationcited by 5
  2. 2024Inferring Behavior-Specific Context Improves Zero-Shot Generalization in Reinforcement Learningcited by 5
  3. 2024Instance Selection for Dynamic Algorithm Configuration with Reinforcement Learning: Improving Generalizationcited by 5
  4. 2025On the Importance of Reward Design in Reinforcement Learning-based Dynamic Algorithm Configuration: A Case Study on OneMax with (1+(λ,λ))-GAcited by 5
  5. 2025Learning to control inexact Benders decomposition via reinforcement learningcited by 5
  6. 2024CANDID DAC: Leveraging Coupled Action Dimensions with Importance Differences in DACcited by 4
  7. 2025Sequential Multi-Agent Dynamic Algorithm Configurationcited by 3
  8. 2025Frozen Layers: Memory-efficient Many-fidelity Hyperparameter Optimizationcited by 2
  9. 2023Discovering Multiple Algorithm Configurationscited by 1
  10. 2024Athanor: Local Search over Abstract Constraint Specificationscited by 1
  11. 2025Adaptive and Robust DBSCAN with Multi-agent Reinforcement Learningcited by 1
  12. 2025Multi-parameter Control for the (1+(λ,λ))-GA on OneMax via Deep Reinforcement Learningcited by 1
  13. 2025Deep Reinforcement Learning for Dynamic Algorithm Configuration: A Case Study on Optimizing OneMax with the (1+(λ,λ))-GAcited by 1
  14. 2026Large Language Models as Optimization Controllers: Adaptive Continuation for SIMP Topology Optimizationcited by 1
  15. 2026AI Training Manager: Bounded Closed-Loop Control of Adaptive Training Recipescited by 1
  16. 2023Stepsize Learning for Policy Gradient Methods in Contextual Markov Decision Processescited by 0
  17. 2023Hybrid Algorithm Selection and Hyperparameter Tuning on Distributed Machine Learning Resources: A Hierarchical Agent-based Approachcited by 0
  18. 2025Graph-Supported Dynamic Algorithm Configuration for Multi-Objective Combinatorial Optimizationcited by 0
  19. 2025Revisiting Learning Rate Controlcited by 0
  20. 2025OPAL: Operator-Programmed Algorithms for Landscape-Aware Black-Box Optimizationcited by 0
  21. 2025Best Practices For Empirical Meta-Algorithmic Research: Guidelines from the COSEAL Research Networkcited by 0
  22. 2026Decoupling Numerical and Structural Parameters: An Empirical Study on Adaptive Genetic Algorithms via Deep Reinforcement Learning for the Large-Scale TSPcited by 0
  23. 2026AutoSiMP: Autonomous Topology Optimization from Natural Language via LLM-Driven Problem Configuration and Adaptive Solver Controlcited by 0
  24. 2026When does learning pay off? A study on DRL-based dynamic algorithm configuration for carbon-aware schedulingcited by 0
  25. 2026Discovering Interpretable Multi-Parameter Control Policies for Evolutionary Algorithms Using Deep Reinforcement Learningcited by 0
  26. 2026NeurGO: Learning to Generate Elite Candidates for Meta-Black-Box Expensive Optimizationcited by 0
  27. 2026Reinforcement learning to choose optimizerscited by 0