2025-01-12-sporring25a.md

title	openreview	abstract	layout	series	publisher	issn	id	month	tex_title	firstpage	lastpage	page	order	cycles	bibtex_author	author	date	address	container-title	volume	genre	issued	pdf	extras
Locally orderless networks	JNxddbPPWt	We present Locally Orderless Networks (LON) and the theoretical foundation that links them to Convolutional Neural Networks (CNN), Scale-space histograms, and measurement theory. The key elements are a regular sampling of the bias and the derivative of the activation function. We compare LON, CNN, and Scale-space histograms on prototypical single-layer networks. We show how LON and CNN can emulate each other and how LON expands the set of functions computable to non-linear functions such as squaring. We demonstrate simple networks that illustrate the improved performance of LON over CNN on simple tasks for estimating the gradient magnitude squared, for regressing shape area and perimeter lengths, and for explainability of individual pixels’ influence on the result.	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	sporring25a	0	Locally orderless networks	239	244	239-244	239	false	Sporring, Jon and Xu, Peidi and Lu, Jiahao and Lauze, Francois Bernard and Darkner, Sune	given family Jon Sporring given family Peidi Xu given family Jiahao Lu given family Francois Bernard Lauze given family Sune Darkner	2025-01-12		Proceedings of the 6th Northern Lights Deep Learning Conference (NLDL)	265	inproceedings	date-parts 2025 1 12	https://raw.githubusercontent.com/mlresearch/v265/main/assets/sporring25a/sporring25a.pdf