Conditional Neural Bayes Ratio Estimation for Experimental Design Optimisation

TL;DR

The paper introduces cNBRE, a neural network method for optimizing experimental designs by estimating Bayes factors across continuous parameter spaces, demonstrated in radio cosmology.

Contribution

It extends neural Bayes ratio estimation to conditional settings, enabling systematic design exploration with a single trained model.

Findings

cNBRE enables exploration of design space in radio cosmology simulations.

It reveals a 20% variation in detection probability based on antenna orientation.

The method recovers known physical relationships in the application.

Abstract

For frontier experiments operating at the edge of detectability, instrument design directly determines the probability of discovery. We introduce Conditional Neural Bayes Ratio Estimation (cNBRE), which extends neural Bayes ratio estimation by conditioning on design parameters, enabling a single trained network to estimate Bayes factors across a continuous design space. Applied to 21-cm radio cosmology with simulations representative of the REACH experiment, the amortised nature of cNBRE enables systematic design space exploration that would be intractable with traditional point-wise methods, while recovering established physical relationships. The analysis demonstrates a ~20 percentage point variation in detection probability with antenna orientation for a single night of observation, a design decision that would be trivial to implement if determined prior to antenna construction. This framework enables efficient, globally-informed experimental design optimisation for a wide range of scientific applications.