Bayesian Optimization Allowing for Common Random Numbers

by   Michael Pearce, et al.

Bayesian optimization is a powerful tool for expensive stochastic black-box optimization problems such as simulation-based optimization or machine learning hyperparameter tuning. Many stochastic objective functions implicitly require a random number seed as input. By explicitly reusing a seed a user can exploit common random numbers, comparing two or more inputs under the same randomly generated scenario, such as a common customer stream in a job shop problem, or the same random partition of training data into training and validation set for a machine learning algorithm. With the aim of finding an input with the best average performance over infinitely many seeds, we propose a novel Gaussian process model that jointly models both the output for each seed and the average. We then introduce the Knowledge gradient for Common Random Numbers that iteratively determines a combination of input and random seed to evaluate the objective and automatically trades off reusing old seeds and querying new seeds, thus overcoming the need to evaluate inputs in batches or measuring differences of pairs as suggested in previous methods. We investigate the Knowledge Gradient for Common Random Numbers both theoretically and empirically, finding it achieves significant performance improvements with only moderate added computational cost.



There are no comments yet.


page 1

page 2

page 3

page 4


Predictive Entropy Search for Multi-objective Bayesian Optimization with Constraints

This work presents PESMOC, Predictive Entropy Search for Multi-objective...

Bayesian Optimization with Expensive Integrands

We propose a Bayesian optimization algorithm for objective functions tha...

Stratified Bayesian Optimization

We consider derivative-free black-box global optimization of expensive n...

On Model Stability as a Function of Random Seed

In this paper, we focus on quantifying model stability as a function of ...

Trusted-Maximizers Entropy Search for Efficient Bayesian Optimization

Information-based Bayesian optimization (BO) algorithms have achieved st...

Federated Bayesian Optimization via Thompson Sampling

Bayesian optimization (BO) is a prominent approach to optimizing expensi...

Optimisation of a Crossdocking Distribution Centre Simulation Model

This paper reports on continuing research into the modelling of an order...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.