Deploying model obfuscation: towards the privacy of decision-making models on shared platforms حسب Sadhukhan, Payel

احصل على النص الكامل
احصل على النص الكامل

New primitives to AOP weaving capabilities for security hardening concerns حسب Mourad, Azzam

احصل على النص الكامل
احصل على النص الكامل
احصل على النص الكامل
احصل على النص الكامل

Distilling Wisdom: A Review on Optimizing Learning From Massive Language Models حسب Dingzong Zhang (23275066)

"…This survey provides a detailed discussion of the basic principles, algorithms, and implementation methods of knowledge distillation. …"

Common weaving approach in mainstream languages for software security hardening حسب Alhadidi, Dima

"…To achieve this goal, we present a formal specification for GIMPLE weaving and the implementation strategies of the proposed weaving semantics. …"
احصل على النص الكامل
احصل على النص الكامل
احصل على النص الكامل

The automation of the development of classification models and improvement of model quality using feature engineering techniques حسب Sjoerd Boeschoten (17347045)

"…The proposed framework is extendable and configurable by adding algorithms supported by the CARET package implemented in the R programming language. …"

Dynamic Cyber Resilience of Interdependent Critical Information Infrastructures حسب JUMA, MAZEN GHAZI

"…The technology stack of the proposed solution was also implemented with four algorithms and eight protocols. …"
احصل على النص الكامل

Towards secure private and trustworthy human-centric embedded machine learning: An emotion-aware facial recognition case study حسب Muhammad Atif Butt (10849980)

"…However, despite the significant impact of AI design on human interests, the security and trustworthiness of edge AI applications are not foolproof and ethicalneither foolproof nor ethical; Moreover, social norms are often ignored duringin the design, implementation, and deployment of edge AI systems. In this paper, we make the following two contributions: Firstly, we analyze the application of edge AI through a human-centric perspective. …"

Wearable wrist to finger photoplethysmogram translation through restoration using super operational neural networks based 1D-CycleGAN for enhancing cardiovascular monitoring حسب Sakib Mahmud (15302404)

"…Current signal processing techniques, and even state-of-the-art machine learning algorithms, frequently struggle to effectively restore the inherent bodily signals amidst the array of randomly generated distortions. …"