# Output Space Entropy Search Framework for Multi-Objective Bayesian Optimization

@article{Belakaria2021OutputSE,
title={Output Space Entropy Search Framework for Multi-Objective Bayesian Optimization},
author={Syrine Belakaria and Aryan Deshwal and Janardhan Rao Doppa},
journal={J. Artif. Intell. Res.},
year={2021},
volume={72},
pages={667-715}
}
• Published 13 October 2021
• Computer Science
• J. Artif. Intell. Res.
We consider the problem of black-box multi-objective optimization (MOO) using expensive function evaluations (also referred to as experiments), where the goal is to approximate the true Pareto set of solutions by minimizing the total resource cost of experiments. For example, in hardware design optimization, we need to find the designs that trade-off performance, energy, and area overhead using expensive computational simulations. The key challenge is to select the sequence of experiments to…

## Figures and Tables from this paper

### Max-value Entropy Search for Multi-objective Bayesian Optimization with Constraints

• Computer Science
ArXiv
• 2020
A Bayesian optimization method that can be used to solve constrained multi-objective problems when the objectives and the constraints are expensive to evaluate, and its execution time is smaller than other information-based methods.

### PF2ES: Parallel Feasible Pareto Frontier Entropy Search for Multi-Objective Bayesian Optimization Under Unknown Constraints

• Computer Science
ArXiv
• 2022
This work presents Parallel Feasible Pareto Frontier Entropy Search ( { PF ES) — a novel information-theoretic acquisition function for multi-objective Bayesian optimization that provides a low cost and accurate estimate of the mutual information for the parallel setting.

### Joint Entropy Search for Multi-objective Bayesian Optimization

• Computer Science
ArXiv
• 2022
This paper proposes a novel information-theoretic acquisition function for BO called Joint Entropy Search (JES), which considers the joint information gain for the optimal set of inputs and outputs and presents several analytical approximations to the JES acquisition function.

### $\{\text{PF}\}^2\text{ES}$: Parallel Feasible Pareto Frontier Entropy Search for Multi-Objective Bayesian Optimization Under Unknown Constraints

• Computer Science
• 2022
This work presents Parallel Feasible Pareto Frontier Entropy Search ( { PF ES) — a novel information-theoretic acquisition function for multi-objective Bayesian optimization that provides a low cost and accurate estimate of the mutual information for the parallel setting.

### Bayesian optimization of nanoporous materials

• Computer Science
Molecular Systems Design & Engineering
• 2021
Nanoporous materials could be used to store, capture, and sense many different gases, and given an adsorption task, one often wishes to search a library of NPMs for the one suitable for the task.

## References

SHOWING 1-10 OF 71 REFERENCES

### Multi-Fidelity Multi-Objective Bayesian Optimization: An Output Space Entropy Search Approach

• Computer Science
AAAI
• 2020
Experiments show that MF-OSEMO, with both approximations, significantly improves over the state-of-the-art single-fidelity algorithms for multi-objective optimization.

### Uncertainty-Aware Search Framework for Multi-Objective Bayesian Optimization

• Computer Science
AAAI
• 2020
This work proposes a novel uncertainty-aware search framework referred to as USeMO to efficiently select the sequence of inputs for evaluation to solve the problem of multi-objective (MO) blackbox optimization using expensive function evaluations.

### Max-value Entropy Search for Multi-objective Bayesian Optimization with Constraints

• Computer Science
ArXiv
• 2020
A Bayesian optimization method that can be used to solve constrained multi-objective problems when the objectives and the constraints are expensive to evaluate, and its execution time is smaller than other information-based methods.

### Parallel Predictive Entropy Search for Multi-objective Bayesian Optimization with Constraints

• Computer Science
ArXiv
• 2020
PPESMOC selects, at each iteration, a batch of input locations at which to evaluate the black-boxes, in parallel, to maximally reduce the entropy of the problem solution, the first batch method for constrained multi-objective BO.

### Multi-fidelity Bayesian Optimization with Max-value Entropy Search

• Computer Science
ICML
• 2020
This work proposes a novel information theoretic approach to multi-fidelity Bayesian optimization (MFBO) based on a variant of information-based BO called max-value entropy search (MES), which greatly facilitates evaluation of the information gain in MFBO.

### ParEGO: a hybrid algorithm with on-line landscape approximation for expensive multiobjective optimization problems

Results show that NSGA-II, a popular multiobjective evolutionary algorithm, performs well compared with random search, even within the restricted number of evaluations used.

### Predictive Entropy Search for Efficient Global Optimization of Black-box Functions

• Computer Science
NIPS
• 2014
This work proposes a novel information-theoretic approach for Bayesian optimization called Predictive Entropy Search (PES), which codifies this intractable acquisition function in terms of the expected reduction in the differential entropy of the predictive distribution.

### Multiobjective Optimization on a Limited Budget of Evaluations Using Model-Assisted -Metric Selection

• Computer Science
PPSN
• 2008
This paper provides a review of contemporary multiobjective approaches based on the singleobjective meta-model-assisted 'Efficient Global Optimization' (EGO) procedure and describes their main concepts and introduces a new EGO-based MOOA, which utilizes the $\mathcal{S}$-metric or hypervolume contribution to decide which solution is evaluated next.

### MUMBO: MUlti-task Max-value Bayesian Optimization

• Computer Science
ECML/PKDD
• 2020
A novel multi-task version of entropy search is derived, delivering robust performance with low computational overheads across classic optimization challenges and multi- task hyper-parameter tuning.

### Optimizing Discrete Spaces via Expensive Evaluations: A Learning to Search Framework

• Computer Science
AAAI
• 2020
The main contribution is to introduce and evaluate a new learning-to-search framework for this problem called L2S-DISCO, to employ search procedures guided by control knowledge at each step to select the next structure and to improve the control knowledge as new function evaluations are observed.