mlrMBO: A Modular Framework for Model-Based Optimization of Expensive Black-Box Functions

Abstract

We present mlrMBO, a flexible and comprehensive R toolbox for model-based optimization (MBO), also known as Bayesian optimization, which addresses the problem of expensive black-box optimization by approximating the given objective function through a surrogate regression model. It is designed for both single- and multi-objective optimization with mixed continuous, categorical and conditional parameters. Additional features include multi-point batch proposal, parallelization, visualization, logging and error-handling. mlrMBO is implemented in a modular fashion, such that single components can be easily replaced or adapted by the user for specific use cases, e.g., any regression learner from the mlr toolbox for machine learning can be used, and infill criteria and infill optimizers are easily exchangeable. We empirically demonstrate that mlrMBO provides state-of-the-art performance by comparing it on different benchmark scenarios against a wide range of other optimizers, including DiceOptim, rBayesianOptimization, SPOT, SMAC, Spearmint, and Hyperopt.

23 pages, 5 figures

Cited by in corpus (20)

• Performance evaluation and hyperparameter tuning of statistical and machine-learning models using spatial data
• Tunability: Importance of Hyperparameters of Machine Learning Algorithms
• Understanding the effect of hyperparameter optimization on machine learning models for structure design problems
• OBOE: Collaborative Filtering for AutoML Model Selection
• Predictive Entropy Search for Multi-objective Bayesian Optimization with Constraints
• OpenML: An R Package to Connect to the Machine Learning Platform OpenML
• Planning as Optimization: Dynamically Discovering Optimal Configurations for Runtime Situations
• XGBoostLSS -- An extension of XGBoost to probabilistic forecasting
• Multi-Objective Automatic Machine Learning with AutoxgboostMC
• HPC Storage Service Autotuning Using Variational-Autoencoder-Guided Asynchronous Bayesian Optimization
• ReinBo: Machine Learning pipeline search and configuration with Bayesian Optimization embedded Reinforcement Learning
• High Dimensional Restrictive Federated Model Selection with multi-objective Bayesian Optimization over shifted distributions
• Bayesian Optimization in Materials Science: A Survey
• A Simulated Annealing Algorithm for Joint Stratification and Sample Allocation Designs
• Fibonacci and k-Subsecting Recursive Feature Elimination
• Empowering individual trait prediction using interactions
• Infill Criterion for Multimodal Model-Based Optimisation
• A First Analysis of Kernels for Kriging-based Optimization in Hierarchical Search Spaces
• varycoef: An R Package for Gaussian Process-based Spatially Varying Coefficient Models
• Modeling and Optimizing Laser-Induced Graphene