Maintaining Adversarial Robustness in Continuous Learning

arXiv:2402.11196v1 Announce Type: new
Abstract: Adversarial robustness is essential for security and reliability of machine learning systems. However, the adversarial robustness gained by sophisticated defense algorithms is easily erased as the neural network evolves to learn new tasks. This vulnerability can be addressed by fostering a novel capability for neural networks, termed continual robust learning, which focuses on both the (classification) performance and adversarial robustness on previous tasks during continuous learning. To achieve continuous robust learning, we propose an approach …

abstract adversarial algorithms arxiv capability continual continuous cs.ai cs.lg defense learn learning systems machine machine learning network networks neural network neural networks novel reliability robust robustness security systems tasks type vulnerability