Combining Different V1 Brain Model Variants to Improve Robustness to Image Corruptions in CNNs

Abstract

While some convolutional neural networks (CNNs) have surpassed human visual abilities in object classification, they often struggle to recognize objects in images corrupted with different types of common noise patterns, highlighting a major limitation of this family of models. Recently, it has been shown that simulating a primary visual cortex (V1) at the front of CNNs leads to small improvements in robustness to these image perturbations. In this study, we start with the observation that different variants of the V1 model show gains for specific corruption types. We then build a new model using an ensembling technique, which combines multiple individual models with different V1 front-end variants. The model ensemble leverages the strengths of each individual model, leading to significant improvements in robustness across all corruption categories and outperforming the base model by 38% on average. Finally, we show that using distillation, it is possible to partially compress the knowledge in the ensemble model into a single model with a V1 front-end. While the ensembling and distillation techniques used here are hardly biologically-plausible, the results presented here demonstrate that by combining the specific strengths of different neuronal circuits in V1 it is possible to improve the robustness of CNNs for a wide range of perturbations.

15 pages with supplementary material, 3 main figures, 2 supplementary figures, 4 supplementary tables

References in corpus (11)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Distilling the Knowledge in a Neural Network
• Popular Ensemble Methods: An Empirical Study
• On Evaluating Adversarial Robustness
• Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning
• Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation
• BatchEnsemble: An Alternative Approach to Efficient Ensemble and Lifelong Learning
• An Effective Anti-Aliasing Approach for Residual Networks
• Diverse Ensembles Improve Calibration
• Noisy Self-Knowledge Distillation for Text Summarization
• Towards robust vision by multi-task learning on monkey visual cortex