ZK Cost-Aware Model Training: An Empirical Analysis of ReLU Activation Count as a Predictor of Zero-Knowledge Proof Cost | Zenodo

Skip to main

You are using an outdated browser. Please upgrade your browser to improve your experience.

Published July 8, 2026

| Version v1

Publication

Open

ZK Cost-Aware Model Training: An Empirical Analysis of ReLU Activation Count as a Predictor of Zero-Knowledge Proof Cost

Authors/Creators

Abdulrahman

Description

Abstract

Zero-knowledge machine learning (zkML) enables cryptographic verification of machine learning inference while preserving privacy, but proof generation remains a significant computational bottleneck. Existing work primarily focuses on reducing proof cost through post-training optimizations, whereas the influence of architectural design choices during model development has received less attention. This work presents an empirical study of the relationship between neural network ReLU activation count and zero-knowledge proof generation cost using the ezkl/Halo2 framework. Across controlled experiments, ReLU activation count exhibits a strong correlation with proof generation time (Pearson r = 0.90) and proof size (r = 0.91), while parameter count is held constant for the primary comparisons. Motivated by these observations, a lightweight proxy metric is proposed to estimate relative proof cost directly from model architecture without executing the proof pipeline. On MNIST, reducing the number of ReLU activations from three to one decreases proof generation time by approximately 6.5% and proof size by 2.3%, while maintaining comparable classification accuracy. These results suggest that ReLU activation count is a useful architectural indicator of proof cost in the ezkl/Halo2 backend and that incorporating proof-cost considerations during architecture selection may improve the efficiency of zkML deployment. The proposed proxy metric provides a fast method for comparing candidate architectures before proof generation.

Files

zk_cost_aware_model_training_report.pdf

Files<br>(359.4 kB)

Name<br>Size

Download all

zk_cost_aware_model_training_report.pdf

md5:a11e5a87f5178d0645c409a3dcc7c0cf

359.4 kB

Preview

Download

Views

Downloads

Show more details

All versions<br>This version

Views

Total views

Downloads

Total downloads

Data volume

Total data volume

359.4 kB<br>359.4 kB

More info on how stats are collected....

Versions

External resources

Indexed in

OpenAIRE

Communities

Details

DOI

DOI Badge

DOI

10.5281/zenodo.21266805

Markdown

[](https://doi.org/10.5281/zenodo.21266805)

reStructuredText

.. image:: https://zenodo.org/badge/DOI/10.5281/zenodo.21266805.svg<br>:target: https://doi.org/10.5281/zenodo.21266805

HTML

Image URL

https://zenodo.org/badge/DOI/10.5281/zenodo.21266805.svg

Target URL

https://doi.org/10.5281/zenodo.21266805

Resource type<br>Publication

Publisher<br>Zenodo

Rights

License

Creative Commons Attribution 4.0 International

The Creative Commons Attribution license allows re-distribution and re-use of a licensed work on the condition that the creator is appropriately credited.

Read more

Citation

Export

Technical metadata

Created

July 8, 2026

Modified

July 8, 2026

Jump up

This site uses cookies. Find out more on how we use cookies

Accept all cookies<br>Accept only essential cookies