Mining gold from implicit models to improve likelihood-free inference
J. Brehmer, G. Louppe, J. Pavez, and K. Cranmer. (2018)cite arxiv:1805.12244Comment: Code available at https://github.com/johannbrehmer/simulator-mining-example . v2: Fixed typos. v3: Expanded discussion, added Lotka-Volterra example. v4: Improved clarity.
Abstract
Simulators often provide the best description of real-world phenomena.
However, they also lead to challenging inverse problems because the density
they implicitly define is often intractable. We present a new suite of
simulation-based inference techniques that go beyond the traditional
Approximate Bayesian Computation approach, which struggles in a
high-dimensional setting, and extend methods that use surrogate models based on
neural networks. We show that additional information, such as the joint
likelihood ratio and the joint score, can often be extracted from simulators
and used to augment the training data for these surrogate models. Finally, we
demonstrate that these new techniques are more sample efficient and provide
higher-fidelity inference than traditional methods.
Description
[1805.12244] Mining gold from implicit models to improve likelihood-free inference
%0 Journal Article
%1 brehmer2018mining
%A Brehmer, Johann
%A Louppe, Gilles
%A Pavez, Juan
%A Cranmer, Kyle
%D 2018
%K approximate bayesian readings
%T Mining gold from implicit models to improve likelihood-free inference
%U http://arxiv.org/abs/1805.12244
%X Simulators often provide the best description of real-world phenomena.
However, they also lead to challenging inverse problems because the density
they implicitly define is often intractable. We present a new suite of
simulation-based inference techniques that go beyond the traditional
Approximate Bayesian Computation approach, which struggles in a
high-dimensional setting, and extend methods that use surrogate models based on
neural networks. We show that additional information, such as the joint
likelihood ratio and the joint score, can often be extracted from simulators
and used to augment the training data for these surrogate models. Finally, we
demonstrate that these new techniques are more sample efficient and provide
higher-fidelity inference than traditional methods.
@article{brehmer2018mining,
abstract = {Simulators often provide the best description of real-world phenomena.
However, they also lead to challenging inverse problems because the density
they implicitly define is often intractable. We present a new suite of
simulation-based inference techniques that go beyond the traditional
Approximate Bayesian Computation approach, which struggles in a
high-dimensional setting, and extend methods that use surrogate models based on
neural networks. We show that additional information, such as the joint
likelihood ratio and the joint score, can often be extracted from simulators
and used to augment the training data for these surrogate models. Finally, we
demonstrate that these new techniques are more sample efficient and provide
higher-fidelity inference than traditional methods.},
added-at = {2020-01-22T06:40:18.000+0100},
author = {Brehmer, Johann and Louppe, Gilles and Pavez, Juan and Cranmer, Kyle},
biburl = {https://www.bibsonomy.org/bibtex/2e4e37b0087a3b381371e9d443adb766d/kirk86},
description = {[1805.12244] Mining gold from implicit models to improve likelihood-free inference},
interhash = {39505a027c49f8bab4102b4bf9e1edbf},
intrahash = {e4e37b0087a3b381371e9d443adb766d},
keywords = {approximate bayesian readings},
note = {cite arxiv:1805.12244Comment: Code available at https://github.com/johannbrehmer/simulator-mining-example . v2: Fixed typos. v3: Expanded discussion, added Lotka-Volterra example. v4: Improved clarity},
timestamp = {2020-01-22T06:40:18.000+0100},
title = {Mining gold from implicit models to improve likelihood-free inference},
url = {http://arxiv.org/abs/1805.12244},
year = 2018
}