Error Aware Distribution Prediction for Lightweight Implicit Neural Representations

Authors

Zhimin Li (Vanderbilt University), Jake Dylan Balla (University of Arizona), Joshua A Levine (University of Arizona)

Presentation

Session

Can We Trust This Chart? (Asking for a Friend)

Time

Wednesday, Nov 11, 08:36 – 08:45 (US/Eastern) · session 08:00 – 09:30

Location

Hall America south

Keywords

Scientific Visualization, Implicit Neural Representation, Uncertainty Quantification

Abstract

Implicit neural representations (INRs) offer compact encoding of volumes, but as lossy approximators, inevitably have prediction errors. We consider INRs that can simultaneously encode relative error scales by predicting distributions using tools from uncertainty estimation. Typically, uncertainty estimation relies on computationally expensive approaches or on predefined parametric assumptions about the predictive distribution (e.g., Gaussian). In this study, we propose a lightweight method that reformulates regression-based INR training as a classification task by discretizing continuous targets into bins, enabling flexible distribution modeling to capture complex multimodal behaviors. We analyze the trade-off between regression and classification for INR training and demonstrate that the classification setting tends to achieve high reconstruction quality and competitive error awareness through uncertainty estimation, compared to regression-based approaches.

For Practitioners

simulation scientists,data scientists