From MNIST to ImageNet: Understanding the Scalability Boundaries of Differentiable Logic Gate Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Brändle, Sven, Aczel, Till, Plesner, Andreas, Wattenhofer, Roger
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909816567365632
author Brändle, Sven
Aczel, Till
Plesner, Andreas
Wattenhofer, Roger
author_facet Brändle, Sven
Aczel, Till
Plesner, Andreas
Wattenhofer, Roger
contents Differentiable Logic Gate Networks (DLGNs) are a very fast and energy-efficient alternative to conventional feed-forward networks. With learnable combinations of logical gates, DLGNs enable fast inference by hardware-friendly execution. Since the concept of DLGNs has only recently gained attention, these networks are still in their developmental infancy, including the design and scalability of their output layer. To date, this architecture has primarily been tested on datasets with up to ten classes. This work examines the behavior of DLGNs on large multi-class datasets. We investigate its general expressiveness, its scalability, and evaluate alternative output strategies. Using both synthetic and real-world datasets, we provide key insights into the importance of temperature tuning and its impact on output layer performance. We evaluate conditions under which the Group-Sum layer performs well and how it can be applied to large-scale classification of up to 2000 classes.
format Preprint
id arxiv_https___arxiv_org_abs_2509_25933
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From MNIST to ImageNet: Understanding the Scalability Boundaries of Differentiable Logic Gate Networks
Brändle, Sven
Aczel, Till
Plesner, Andreas
Wattenhofer, Roger
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Differentiable Logic Gate Networks (DLGNs) are a very fast and energy-efficient alternative to conventional feed-forward networks. With learnable combinations of logical gates, DLGNs enable fast inference by hardware-friendly execution. Since the concept of DLGNs has only recently gained attention, these networks are still in their developmental infancy, including the design and scalability of their output layer. To date, this architecture has primarily been tested on datasets with up to ten classes. This work examines the behavior of DLGNs on large multi-class datasets. We investigate its general expressiveness, its scalability, and evaluate alternative output strategies. Using both synthetic and real-world datasets, we provide key insights into the importance of temperature tuning and its impact on output layer performance. We evaluate conditions under which the Group-Sum layer performs well and how it can be applied to large-scale classification of up to 2000 classes.
title From MNIST to ImageNet: Understanding the Scalability Boundaries of Differentiable Logic Gate Networks
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.25933