Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cannistraci, Irene, Moschella, Luca, Fumero, Marco, Maiorca, Valentino, Rodolà, Emanuele
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2310.01211
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914721349763072
author Cannistraci, Irene
Moschella, Luca
Fumero, Marco
Maiorca, Valentino
Rodolà, Emanuele
author_facet Cannistraci, Irene
Moschella, Luca
Fumero, Marco
Maiorca, Valentino
Rodolà, Emanuele
contents It has been observed that representations learned by distinct neural networks conceal structural similarities when the models are trained under similar inductive biases. From a geometric perspective, identifying the classes of transformations and the related invariances that connect these representations is fundamental to unlocking applications, such as merging, stitching, and reusing different neural modules. However, estimating task-specific transformations a priori can be challenging and expensive due to several factors (e.g., weights initialization, training hyperparameters, or data modality). To this end, we introduce a versatile method to directly incorporate a set of invariances into the representations, constructing a product space of invariant components on top of the latent representations without requiring prior knowledge about the optimal invariance to infuse. We validate our solution on classification and reconstruction tasks, observing consistent latent similarity and downstream performance improvements in a zero-shot stitching setting. The experimental analysis comprises three modalities (vision, text, and graphs), twelve pretrained foundational models, nine benchmarks, and several architectures trained from scratch.
format Preprint
id arxiv_https___arxiv_org_abs_2310_01211
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle From Bricks to Bridges: Product of Invariances to Enhance Latent Space Communication
Cannistraci, Irene
Moschella, Luca
Fumero, Marco
Maiorca, Valentino
Rodolà, Emanuele
Machine Learning
It has been observed that representations learned by distinct neural networks conceal structural similarities when the models are trained under similar inductive biases. From a geometric perspective, identifying the classes of transformations and the related invariances that connect these representations is fundamental to unlocking applications, such as merging, stitching, and reusing different neural modules. However, estimating task-specific transformations a priori can be challenging and expensive due to several factors (e.g., weights initialization, training hyperparameters, or data modality). To this end, we introduce a versatile method to directly incorporate a set of invariances into the representations, constructing a product space of invariant components on top of the latent representations without requiring prior knowledge about the optimal invariance to infuse. We validate our solution on classification and reconstruction tasks, observing consistent latent similarity and downstream performance improvements in a zero-shot stitching setting. The experimental analysis comprises three modalities (vision, text, and graphs), twelve pretrained foundational models, nine benchmarks, and several architectures trained from scratch.
title From Bricks to Bridges: Product of Invariances to Enhance Latent Space Communication
topic Machine Learning
url https://arxiv.org/abs/2310.01211