Architectural changes boost image model defense without costly training nine words
Neither Adversarial Training Nor Purification: Emergent Adversarial Robustness from Oscillatory Predictive Learning
Machine LearningArtificial Intelligence
Summary
Making computers recognize images correctly even when someone tries to trick them is usually slow and expensive because it involves special training or cleaning steps. This paper shows that careful design of the model and the way it learns can naturally make it harder to fool. The authors created a new method called Oscillatory Predictive Learning that uses special neuron designs and self-teaching to improve robustness. Their approach works well on standard image tests without needing the usual time-consuming defense methods.
adversarial robustnessadversarial trainingpredictive self-supervised learningKuramoto oscillatorsCIFAR-10CIFAR-100AutoAttackneural networkstest-time purification
Authors
Mohammed-Yassine Habibi, Klea Ziu, Martin Takáč, Makoto Yamada
Abstract
Adversarial robustness in computer vision is still largely achieved through adversarial training or test-time adversarial purification, both of which introduce significant computational overhead by generating adversarial examples during training or performing iterative denoising at test time. We study whether empirical robustness can instead emerge from architectural and representation-learning inductive biases. We introduce Oscillatory Predictive Learning (OPL), a two-stage framework that combines Artificial Kuramoto Oscillatory Neurons (AKOrN) with predictive self-supervised pretraining using X-PhiNet. Because our default checkpoint uses randomized initial oscillator states, we compare it with other randomized adversarial defense methods that provide precise, reproducible, and strong attack protocols. Experiments on CIFAR-10 and CIFAR-100, with additional corruption evaluation on CIFAR-10-C, demonstrate that our method achieves competitive results under the AutoAttack-rand evaluation protocol. On CIFAR-10 and CIFAR-100, OPL attains 76.63$\pm$0.76$\%$ and 50.44$\%$ robust accuracy, respectively, under $\ell_\infty$, $ε=8/255$, AutoAttack-rand with EoT $K=20$.