**arXiv ID:** 2502.16003 **Authors:** Francisco M. López, Jochen Triesch **Published:** 2025-02-21T23:35:08Z **Abstract:** We present Hierarchical Residual Networks (HiResNets), deep convolutional neural networks with long-range residual connections between layers at different hierarchical levels. HiResNets draw inspiration on the organization of the mammalian brain by replicating the direct connections from subcortical areas to the entire cortical hierarchy. We show that the inclusion of hie...
Installs into .claude/skills of the current project.
Sponsors
SSkills DirectorySkills Directory
Know which skills are safe — weekly.
SSkills DirectorySkills Directory
Is that skill safe to install?
Are you the author of Hierarchical Residuals Exploit Braininspired Compositionality?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/hiyenwong-hierarchical-residuals-exploit-braininspired-compo)
# Hierarchical Residuals Exploit Brain-Inspired Compositionality
**arXiv ID:** 2502.16003
**Authors:** Francisco M. López, Jochen Triesch
**Published:** 2025-02-21T23:35:08Z
**Abstract:**
We present Hierarchical Residual Networks (HiResNets), deep convolutional neural networks with long-range residual connections between layers at different hierarchical levels. HiResNets draw inspiration on the organization of the mammalian brain by replicating the direct connections from subcortical areas to the entire cortical hierarchy. We show that the inclusion of hierarchical residuals in several architectures, including ResNets, results in a boost in accuracy and faster learning. A detailed analysis of our models reveals that they perform hierarchical compositionality by learning feature maps relative to the compressed representations provided by the skip connections.
## Skill Description
This skill is generated from the arXiv paper: Hierarchical Residuals Exploit Brain-Inspired Compositionality (2502.16003).
## How to Use
[To be filled in by the user or by future automation]
## References
- [arXiv:2502.16003](http://arxiv.org/abs/2502.16003v1)