Guiding the retraining of convolutional neural networks against adversarial inputs

التفاصيل البيبلوغرافية
العنوان:	Guiding the retraining of convolutional neural networks against adversarial inputs
المؤلفون:	Durán López, Francisco Javier, Martínez Fernández, Silverio Juan, Felderer, Michael, Franch Gutiérrez, Javier
المساهمون:	Universitat Politècnica de Catalunya. Departament d'Enginyeria de Serveis i Sistemes d'Informació, Universitat Politècnica de Catalunya. inSSIDE - integrated Software, Services, Information and Data Engineering
بيانات النشر:	PeerJ
سنة النشر:	2023
المجموعة:	Universitat Politècnica de Catalunya, BarcelonaTech: UPCommons - Global access to UPC knowledge
مصطلحات موضوعية:	Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic, Computer software -- Testing, Deep learning, Energy consumption, Neural networks, Software testing, Adversarial inputs, Green AI, Programari -- Tests, Aprenentatge profund, Energia -- Consum
الوصف:	Background: When using deep learning models, one of the most critical vulnerabilities is their exposure to adversarial inputs, which can cause wrong decisions (e.g., incorrect classification of an image) with minor perturbations. To address this vulnerability, it becomes necessary to retrain the affected model against adversarial inputs as part of the software testing process. In order to make this process energy efficient, data scientists need support on which are the best guidance metrics for reducing the adversarial inputs to create and use during testing, as well as optimal dataset configurations. Aim: We examined six guidance metrics for retraining deep learning models, specifically with convolutional neural network architecture, and three retraining configurations. Our goal is to improve the convolutional neural networks against the attack of adversarial inputs with regard to the accuracy, resource utilization and execution time from the point of view of a data scientist in the context of image classification. Method: We conducted an empirical study using five datasets for image classification. We explore: (a) the accuracy, resource utilization, and execution time of retraining convolutional neural networks with the guidance of six different guidance metrics (neuron coverage, likelihood-based surprise adequacy, distance-based surprise adequacy, DeepGini, softmax entropy and random), (b) the accuracy and resource utilization of retraining convolutional neural networks with three different configurations (one-step adversarial retraining, adversarial retraining and adversarial fine-tuning). Results: We reveal that adversarial retraining from original model weights, and by ordering with uncertainty metrics, gives the best model w.r.t. accuracy, resource utilization, and execution time. Conclusions: Although more studies are necessary, we recommend data scientists use the above configuration and metrics to deal with the vulnerability to adversarial inputs of deep learning models, as they can improve their ...
نوع الوثيقة:	article in journal/newspaper
وصف الملف:	31 p.; application/pdf
اللغة:	English
تدمد:	2376-5992
Relation:	https://peerj.com/articles/cs-1454/; info:eu-repo/grantAgreement/AEI/PLAN ESTATAL DE INVESTIGACIÓN CIENTÍFICA Y TÉCNICA Y DE INNOVACIÓN 2021-2023/TED2021-130923B-I00/GAISSA. Transición hacia sistemas de software verdes basados en IA: un enfoque centrado en arquitectura; Duran, F. [et al.]. Guiding the retraining of convolutional neural networks against adversarial inputs. "PeerJ. Computer science", 8 Agost 2023, vol. 9, article 1454.; https://arxiv.org/abs/2207.03689; http://hdl.handle.net/2117/395100
DOI:	10.7717/peerj-cs.1454
الاتاحة:	http://hdl.handle.net/2117/395100 https://arxiv.org/abs/2207.03689 https://doi.org/10.7717/peerj-cs.1454
Rights:	Attribution 4.0 International ; http://creativecommons.org/licenses/by/4.0/ ; Open Access
رقم الانضمام:	edsbas.EFDA093A
قاعدة البيانات:	BASE

View record in BASE

Full Text Finder

الوصف
تدمد:	23765992
DOI:	10.7717/peerj-cs.1454