The CMA Evolution Strategy: A Tutorial

Abstract

This tutorial introduces the CMA Evolution Strategy (ES), where CMA stands for Covariance Matrix Adaptation. The CMA-ES is a stochastic, or randomized, method for real-parameter (continuous domain) optimization of non-linear, non-convex functions. We try to motivate and derive the algorithm from intuitive concepts and from requirements of non-linear, non-convex search in continuous domain.

ArXiv e-prints, arXiv:1604.00772, 2016, pp.1-39

Cited by in corpus (46)

• GCN-RL Circuit Designer: Transferable Transistor Sizing with Graph Neural Networks and Reinforcement Learning
• A determination of the fragmentation functions of pions, kaons, and protons with faithful uncertainties
• A review on deep reinforcement learning for fluid mechanics: an update
• A review on Deep Reinforcement Learning for Fluid Mechanics
• Classical Optimizers for Noisy Intermediate-Scale Quantum Devices
• Graph neural networks for simulating crack coalescence and propagation in brittle materials
• OpenFPM: A scalable open framework for particle and particle-mesh codes on parallel computers
• CMA-ES with Margin: Lower-Bounding Marginal Probability for Mixed-Integer Black-Box Optimization
• The Free Energy Principle for Perception and Action: A Deep Learning Perspective
• Optimized and autonomous machine learning framework for characterizing pores, particles, grains and grain boundaries in microstructural images
• LEAP: Scaling Numerical Optimization Based Synthesis Using an Incremental Approach
• A Population-based Hybrid Approach to Hyperparameter Optimization for Neural Networks
• Training Passive Photonic Reservoirs with Integrated Optical Readout
• Predicting Game Difficulty and Churn Without Players
• MMES: Mixture Model based Evolution Strategy for Large-Scale Optimization
• DAMPE squib? Significance of the 1.4 TeV DAMPE excess
• Empirical analysis of PGA-MAP-Elites for Neuroevolution in Uncertain Domains
• Benchmarking Feature-based Algorithm Selection Systems for Black-box Numerical Optimization
• CMA-ES with Learning Rate Adaptation: Can CMA-ES with Default Population Size Solve Multimodal and Noisy Problems?
• Rapid Estimation of Left Ventricular Contractility with a Physics-Informed Neural Network Inverse Modeling Approach
• MCNNTUNES: tuning Shower Monte Carlo generators with machine learning
• DynamoRep: Trajectory-Based Population Dynamics for Classification of Black-box Optimization Problems
• Direct estimation of minimum gate fidelity
• Multi-variable integration with a variational quantum circuit
• Transfer Learning based Search Space Design for Hyperparameter Tuning
• Digging its own Site: Linear Coordination Stabilizes a Pt1/Fe2O3 Single-Atom Catalyst
• Quality-Diversity Meta-Evolution: customising behaviour spaces to a meta-objective
• Distributed Evolution Strategies with Multi-Level Learning for Large-Scale Black-Box Optimization
• On Entropy Regularized Path Integral Control for Trajectory Optimization
• Adapting JPEG XS gains and priorities to tasks and contents
• Structure Parameter Optimized Kernel Based Online Prediction with a Generalized Optimization Strategy for Nonstationary Time Series
• A Modularized Design Approach for GelSight Family of Vision-based Tactile Sensors
• Adaptive Ranking Based Constraint Handling for Explicitly Constrained Black-Box Optimization
• Exploring Public Attention in the Circular Economy through Topic Modelling with Twin Hyperparameter Optimisation
• Generating and Adapting to Diverse Ad-Hoc Cooperation Agents in Hanabi
• A Framework for Population-Based Stochastic Optimization on Abstract Riemannian Manifolds
• Analysis of the $(\mu /{\mu }_I,\lambda )$-CSA-ES with Repair by Projection Applied to a Conically Constrained Problem
• Mathematical modelling, selection and hierarchical inference to determine the minimal dose in IFN$\alpha$ therapy against Myeloproliferative Neoplasms
• Skill Generalization with Verbs
• Grammar-based Ordinary Differential Equation Discovery
• On-line Search History-assisted Restart Strategy for Covariance Matrix Adaptation Evolution Strategy
• Managing Expectations and Imbalanced Training Data in Reactive Force Field Development: an Application to Water Adsorption on Alumina
• Self-building Neural Networks
• Searching Search Spaces: Meta-evolving a Geometric Encoding for Neural Networks
• Learning in Sparse Rewards settings through Quality-Diversity algorithms
• Design and characterization of all two-dimensional fragile topological bands