• Corpus ID: 221640588

Bayesian Screening: Multi-test Bayesian Optimization Applied to in silico Material Screening

@article{Hook2020BayesianSM,
  title={Bayesian Screening: Multi-test Bayesian Optimization Applied to in silico Material Screening},
  author={James Hook and Calum Hand and Emma Whitfield},
  journal={ArXiv},
  year={2020},
  volume={abs/2009.05418}
}
We present new multi-test Bayesian optimization models and algorithms for use in large scale material screening applications. Our screening problems are designed around two tests, one expensive and one cheap. This paper differs from other recent work on multi-test Bayesian optimization through use of a flexible model that allows for complex, non-linear relationships between the cheap and expensive test scores. This additional modeling flexibility is essential in the material screening… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 19 REFERENCES

Bayesian optimization for materials design

TLDR
Two Bayesian optimization methods are introduced: expected improvement, for design problems with noise-free evaluations; and the knowledge-gradient method, which generalizes expected improvement and may be used in design Problems with noisy evaluations, and enjoy one-step Bayes-optimality.

Constrained Bayesian Optimization with Noisy Experiments

TLDR
Simulations with synthetic functions show that optimization performance on noisy, constrained problems outperforms existing methods and derive an expression for expected improvement under greedy batch optimization with noisy observations and noisy constraints, and develop a quasi-Monte Carlo approximation.

Active learning in materials science with emphasis on adaptive sampling using uncertainties for targeted design

TLDR
How methods from the information sciences enable us to accelerate the search and discovery of new materials is reviewed and active learning allows us to effectively navigate the search space iteratively to identify promising candidates for guiding experiments and computations.

A General Framework for Multi-fidelity Bayesian Optimization with Gaussian Processes

TLDR
This paper proposes MF-MI-Greedy, a principled algorithmic framework for addressing multi-fidelity Bayesian optimization with complex structural dependencies among multiple outputs, and proposes a simple notion of regret which incorporates the cost of different fidelities.

Practical Multi-fidelity Bayesian Optimization for Hyperparameter Tuning

TLDR
A new acquisition function, the trace-aware knowledge-gradient, is introduced, which efficiently leverages both multiple continuous fidelity controls and trace observations, and outperforms state-of-the-art alternatives for hyperparameter tuning of deep neural networks and large-scale kernel learning.

Parallel and Distributed Thompson Sampling for Large-scale Accelerated Exploration of Chemical Space

TLDR
This work proposes a scalable solution based on a parallel and distributed implementation of Thompson sampling (PDTS), and shows that, in small scale problems, PDTS performs similarly as parallel expected improvement (EI), a batch version of the most widely used BO heuristic.

Gaussian Process Bandit Optimisation with Multi-fidelity Evaluations

TLDR
This work develops \mfgpucb, a novel method based on upper confidence bound techniques that outperforms such naive strategies and other multi-fidelity methods on several synthetic and real experiments and achieves better regret than strategies which ignore multi- fidelity information.

Parallelised Bayesian Optimisation via Thompson Sampling

TLDR
This work design and analyse variations of the classical Thompson sampling procedure for Bayesian optimisation (BO) in settings where function evaluations are expensive but can be performed in parallel and shows that asynchronous TS outperforms a suite of existing parallel BO algorithms in simulations and in an application involving tuning hyper-parameters of a convolutional neural network.

Large-Scale Quantitative Structure–Property Relationship (QSPR) Analysis of Methane Storage in Metal–Organic Frameworks

Metal–organic frameworks (MOFs) present a combinatorial design challenge. The structural building blocks of MOFs can be combined to synthesize a nearly infinite number of materials. This suggests

The high-throughput highway to computational materials design.

TLDR
A current snapshot of high-throughput computational materials design is provided, and the challenges and opportunities that lie ahead are highlighted.