ABC-CDE: Towards Approximate Bayesian Computation with Complex High-Dimensional Data and Limited Simulations
arXiv:1805.05480 · doi:10.1080/10618600.2018.1546594
Abstract
Approximate Bayesian Computation (ABC) is typically used when the likelihood is either unavailable or intractable but where data can be simulated under different parameter settings using a forward model. Despite the recent interest in ABC, high-dimensional data and costly simulations still remain a bottleneck in some applications. There is also no consensus as to how to best assess the performance of such methods without knowing the true posterior. We show how a nonparametric conditional density estimation (CDE) framework, which we refer to as ABC-CDE, help address three nontrivial challenges in ABC: (i) how to efficiently estimate the posterior distribution with limited simulations and different types of data, (ii) how to tune and compare the performance of ABC and related methods in estimating the posterior itself, rather than just certain properties of the density, and (iii) how to efficiently choose among a large set of summary statistics based on a CDE surrogate loss. We provide theoretical and empirical evidence that justify ABC-CDE procedures that {\em directly} estimate and assess the posterior based on an initial ABC sample, and we describe settings where standard ABC and regression-based approaches are inadequate.
References in corpus (20)
- XGBoost: A Scalable Tree Boosting System
- KiDS-450: Cosmological parameter constraints from tomographic weak gravitational lensing
- Adaptive approximate Bayesian computation
- Non-linear regression models for Approximate Bayesian Computation
- Weak Gravitational Lensing and its Cosmological Applications
- Weak lensing for precision cosmology
- A Comparative Review of Dimension Reduction Methods in Approximate Bayesian Computation
- Cosmology with Weak Lensing Surveys
- Cosmology from Cosmic Shear with DES Science Verification Data
- Fast -free Inference of Simulation Models with Bayesian Conditional Density Estimation
- Approximate Bayesian Computation for Astronomical Model Analysis: A Case Study in Galaxy Demographics and Morphological Transformation at High Redshift
- Likelihood-Free Cosmological Inference with Type Ia Supernovae: Approximate Bayesian Computation for a Complete Treatment of Uncertainty
- Flexible statistical inference for mechanistic models of neural dynamics
- Simulations of Weak Gravitational Lensing - II : Including Finite Support Effects in Cosmic Shear Covariance Matrices
- Tailor-made tests for goodness of fit to semiparametric hypotheses
- Mocking the Weak Lensing universe: the LensTools python computing package
- Simulations of Wide-Field Weak Lensing Surveys II: Covariance Matrix of Real Space Correlation Functions
- A Spectral Series Approach to High-Dimensional Nonparametric Regression
- RFCDE: Random Forests for Conditional Density Estimation
- Extending approximate Bayesian computation methods to high dimensions via a Gaussian copula model
Cited by in corpus (15)
- The frontier of simulation-based inference
- Conditional Density Estimation Tools in Python and R with Applications to Photometric Redshifts and Likelihood-Free Cosmological Inference
- A Survey of Bayesian Statistical Approaches for Big Data
- Accurate X-ray Timing in the Presence of Systematic Biases With Simulation-Based Inference
- A Bayesian Optimization approach for calibrating large-scale activity-based transport models
- Marginally-calibrated deep distributional regression
- Diagnostics for Conditional Density Models and Bayesian Inference Algorithms
- Calibration of a bumble bee foraging model using Approximate Bayesian Computation
- Privacy-preserving Federated Bayesian Learning of a Generative Model for Imbalanced Classification of Clinical Data
- (f)RFCDE: Random Forests for Conditional Density Estimation and Functional Data
- Confidence Sets and Hypothesis Testing in a Likelihood-Free Inference Setting
- A review of Approximate Bayesian Computation methods via density estimation: inference for simulator-models
- GRAFFL: Gradient-free Federated Learning of a Bayesian Generative Model
- Detecting conflicting summary statistics in likelihood-free inference
- OCDE: Odds Conditional Density Estimator