Deep Companion Learning: Enhancing Generalization Through Historical Consistency
This addresses the challenge of generalization for deep learning practitioners, but it appears incremental as it builds on existing training methods with a novel consistency-based approach.
The paper tackles the problem of improving generalization in deep neural networks by introducing Deep Companion Learning (DCL), a training method that penalizes inconsistent predictions using historical model versions, resulting in state-of-the-art performance on benchmark datasets like CIFAR-100, Tiny-ImageNet, and ImageNet-1K.
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.