Neural Coherence : Find higher performance to out-of-distribution tasks from few samples

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Guiroy, Simon, Richter, Mats, Chandar, Sarath, Pal, Christopher
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908695134208000
author Guiroy, Simon
Richter, Mats
Chandar, Sarath
Pal, Christopher
author_facet Guiroy, Simon
Richter, Mats
Chandar, Sarath
Pal, Christopher
contents To create state-of-the-art models for many downstream tasks, it has become common practice to fine-tune a pre-trained large vision model. However, it remains an open question of how to best determine which of the many possible model checkpoints resulting from a large training run to use as the starting point. This becomes especially important when data for the target task of interest is scarce, unlabeled and out-of-distribution. In such scenarios, common methods relying on in-distribution validation data become unreliable or inapplicable. This work proposes a novel approach for model selection that operates reliably on just a few unlabeled examples from the target task. Our approach is based on a novel concept: Neural Coherence, which entails characterizing a model's activation statistics for source and target domains, allowing one to define model selection methods with high data-efficiency. We provide experiments where models are pre-trained on ImageNet1K and examine target domains consisting of Food-101, PlantNet-300K and iNaturalist. We also evaluate it in many meta-learning settings. Our approach significantly improves generalization across these different target domains compared to established baselines. We further demonstrate the versatility of Neural Coherence as a powerful principle by showing its effectiveness in training data selection.
format Preprint
id arxiv_https___arxiv_org_abs_2512_05880
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Neural Coherence : Find higher performance to out-of-distribution tasks from few samples
Guiroy, Simon
Richter, Mats
Chandar, Sarath
Pal, Christopher
Machine Learning
Artificial Intelligence
To create state-of-the-art models for many downstream tasks, it has become common practice to fine-tune a pre-trained large vision model. However, it remains an open question of how to best determine which of the many possible model checkpoints resulting from a large training run to use as the starting point. This becomes especially important when data for the target task of interest is scarce, unlabeled and out-of-distribution. In such scenarios, common methods relying on in-distribution validation data become unreliable or inapplicable. This work proposes a novel approach for model selection that operates reliably on just a few unlabeled examples from the target task. Our approach is based on a novel concept: Neural Coherence, which entails characterizing a model's activation statistics for source and target domains, allowing one to define model selection methods with high data-efficiency. We provide experiments where models are pre-trained on ImageNet1K and examine target domains consisting of Food-101, PlantNet-300K and iNaturalist. We also evaluate it in many meta-learning settings. Our approach significantly improves generalization across these different target domains compared to established baselines. We further demonstrate the versatility of Neural Coherence as a powerful principle by showing its effectiveness in training data selection.
title Neural Coherence : Find higher performance to out-of-distribution tasks from few samples
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2512.05880