On the Internal Representations of Graph Metanetworks
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910871756734464 |
|---|---|
| author | Yeom, Taesun Lee, Jaeho |
| author_facet | Yeom, Taesun Lee, Jaeho |
| contents | Weight space learning is an emerging paradigm in the deep learning community. The primary goal of weight space learning is to extract informative features from a set of parameters using specially designed neural networks, often referred to as \emph{metanetworks}. However, it remains unclear how these metanetworks learn solely from parameters. To address this, we take the first step toward understanding \emph{representations} of metanetworks, specifically graph metanetworks (GMNs), which achieve state-of-the-art results in this field, using centered kernel alignment (CKA). Through various experiments, we reveal that GMNs and general neural networks (\textit{e.g.,} multi-layer perceptrons (MLPs) and convolutional neural networks (CNNs)) differ in terms of their representation space. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_09120 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | On the Internal Representations of Graph Metanetworks Yeom, Taesun Lee, Jaeho Machine Learning Artificial Intelligence Weight space learning is an emerging paradigm in the deep learning community. The primary goal of weight space learning is to extract informative features from a set of parameters using specially designed neural networks, often referred to as \emph{metanetworks}. However, it remains unclear how these metanetworks learn solely from parameters. To address this, we take the first step toward understanding \emph{representations} of metanetworks, specifically graph metanetworks (GMNs), which achieve state-of-the-art results in this field, using centered kernel alignment (CKA). Through various experiments, we reveal that GMNs and general neural networks (\textit{e.g.,} multi-layer perceptrons (MLPs) and convolutional neural networks (CNNs)) differ in terms of their representation space. |
| title | On the Internal Representations of Graph Metanetworks |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2503.09120 |