English

Deep Companion Learning: Enhancing Generalization Through Historical Consistency

Computer Vision and Pattern Recognition 2024-07-29 v1 Machine Learning

Abstract

We propose Deep Companion Learning (DCL), a novel training method for Deep Neural Networks (DNNs) that enhances generalization by penalizing inconsistent model predictions compared to its historical performance. To achieve this, we train a deep-companion model (DCM), by using previous versions of the model to provide forecasts on new inputs. This companion model deciphers a meaningful latent semantic structure within the data, thereby providing targeted supervision that encourages the primary model to address the scenarios it finds most challenging. We validate our approach through both theoretical analysis and extensive experimentation, including ablation studies, on a variety of benchmark datasets (CIFAR-100, Tiny-ImageNet, ImageNet-1K) using diverse architectural models (ShuffleNetV2, ResNet, Vision Transformer, etc.), demonstrating state-of-the-art performance.

Keywords

Cite

@article{arxiv.2407.18821,
  title  = {Deep Companion Learning: Enhancing Generalization Through Historical Consistency},
  author = {Ruizhao Zhu and Venkatesh Saligrama},
  journal= {arXiv preprint arXiv:2407.18821},
  year   = {2024}
}

Comments

ECCV 2024

R2 v1 2026-06-28T17:54:44.722Z