Meta-Learning with Evolutionary Strategy for Resilience Optimization of Image Recognition System

• Main Authors: Moskalenko, Viacheslav, Korobov, Artem, Kudravcev, Anton, Boiko, Maxym
• Format: Conference Proceeding
• Language: English
• Subjects: Adaptation models; adversarial attacks; Data acquisition; evolutionary strategies; fault injection; Fault tolerance; Fault tolerant systems; few-shot learning; Image recognition; meta-learning; Metalearning; parameter-efficient transfer learning; resilience; robustness; Training
• ISSN: 2770-4254
• DOI: 10.1109/IDAACS58523.2023.10348942

The problem of optimizing the resilience of image recognition systems to destructive disturbances has not yet been fully solved and is quite relevant for safety- critical applications. The task of optimizing the resilience of image recognition system to disturbing influences is a high- level task in relation to accuracy optimization, which determines the prospects of using the ideas and methods of meta-learning to solve it. Stated research goal is to develop an architectural add-ons and the meta-learning method for optimizing the resilience of an image recognition system to destructive disturbances. Meta-updating with evolutionary strategy is proposed for direct maximization of the expected value of resilience criterion. The experiments were conducted on a model with the ResNet-18 architecture, with an add-on in the form of convolutional adapters and meta- adapters for parallel correction of frozen pretrained modules. It has been experimentally confirmed that the proposed method provides a better resilience to random bit- flipinjection compared to training with fault injection by an average of 5.32%. Also, the proposed method provides a better resilience to adversarial evasion attacks compared to adversarial training by an average of 5.42%.