xInv: Explainable Optimization of Inverse Problems

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Memery, Sean, Denamganai, Kevin, Kapron-King, Anna, Subr, Kartic
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915354499874816
author Memery, Sean
Denamganai, Kevin
Kapron-King, Anna
Subr, Kartic
author_facet Memery, Sean
Denamganai, Kevin
Kapron-King, Anna
Subr, Kartic
contents Inverse problems are central to a wide range of fields, including healthcare, climate science, and agriculture. They involve the estimation of inputs, typically via iterative optimization, to some known forward model so that it produces a desired outcome. Despite considerable development in the explainability and interpretability of forward models, the iterative optimization of inverse problems remains largely cryptic to domain experts. We propose a methodology to produce explanations, from traces produced by an optimizer, that are interpretable by humans at the abstraction of the domain. The central idea in our approach is to instrument a differentiable simulator so that it emits natural language events during its forward and backward passes. In a post-process, we use a Language Model to create an explanation from the list of events. We demonstrate the effectiveness of our approach with an illustrative optimization problem and an example involving the training of a neural network.
format Preprint
id arxiv_https___arxiv_org_abs_2506_11056
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle xInv: Explainable Optimization of Inverse Problems
Memery, Sean
Denamganai, Kevin
Kapron-King, Anna
Subr, Kartic
Machine Learning
Artificial Intelligence
I.2.7
Inverse problems are central to a wide range of fields, including healthcare, climate science, and agriculture. They involve the estimation of inputs, typically via iterative optimization, to some known forward model so that it produces a desired outcome. Despite considerable development in the explainability and interpretability of forward models, the iterative optimization of inverse problems remains largely cryptic to domain experts. We propose a methodology to produce explanations, from traces produced by an optimizer, that are interpretable by humans at the abstraction of the domain. The central idea in our approach is to instrument a differentiable simulator so that it emits natural language events during its forward and backward passes. In a post-process, we use a Language Model to create an explanation from the list of events. We demonstrate the effectiveness of our approach with an illustrative optimization problem and an example involving the training of a neural network.
title xInv: Explainable Optimization of Inverse Problems
topic Machine Learning
Artificial Intelligence
I.2.7
url https://arxiv.org/abs/2506.11056