Computer Engineering Commons^™

The Role Of Attention Mechanisms In Enhancing Transparency And Interpretability Of Neural Network Models In Explainable Ai, Bhargav Kotipalli

Harrisburg University Dissertations and Theses

In the rapidly evolving field of artificial intelligence (AI), deep learning models' interpretability

and reliability are severely hindered by their complexity and opacity. Enhancing the

transparency and interpretability of AI systems for humans is the primary objective of the

emerging field of explainable AI (XAI). The attention mechanisms at the heart of XAI's work

are based on human cognitive processes. Neural networks can now dynamically focus on

relevant parts of the input data thanks to these mechanisms, which enhances interpretability

and performance. This report covers in-depth talks of attention mechanisms in neural networks

within XAI, as well as an analysis …

Spotlight Report #6: Proffering Machine-Readable Personal Privacy Research Agreements: Pilot Project Findings For Ieee P7012 Wg, Noreen Y. Whysel, Lisa Levasseur

Publications and Research

What if people had the ability to assert their own legally binding permissions for data collection, use, sharing, and retention by the technologies they use? The IEEE P7012 has been working on an interoperability specification for machine-readable personal privacy terms to support this ability since 2018. The premise behind the work of IEEE P7012 is that people need technology that works on their behalf—i.e. software agents that assert the individual’s permissions and preferences in a machine-readable format.

Thanks to a grant from the IEEE Technical Activities Board Committee on Standards (TAB CoS), we were able to explore the attitudes of …

Privacy-Preserving Cloud-Assisted Data Analytics, Wei Bao

Graduate Theses and Dissertations

Nowadays industries are collecting a massive and exponentially growing amount of data that can be utilized to extract useful insights for improving various aspects of our life. Data analytics (e.g., via the use of machine learning) has been extensively applied to make important decisions in various real world applications. However, it is challenging for resource-limited clients to analyze their data in an efficient way when its scale is large. Additionally, the data resources are increasingly distributed among different owners. Nonetheless, users' data may contain private information that needs to be protected.

Cloud computing has become more and more popular in …

Multi-Scale, Class-Generic, Privacy-Preserving Video, Zhixiang Zhang, Thomas Cilloni, Charles Walter, Charles Fleming

Faculty and Student Publications

In recent years, high-performance video recording devices have become ubiquitous, posing an unprecedented challenge to preserving personal privacy. As a result, privacy-preserving video systems have been receiving increased attention. In this paper, we present a novel privacy-preserving video algorithm that uses semantic segmentation to identify regions of interest, which are then anonymized with an adaptive blurring algorithm. This algorithm addresses two of the most important shortcomings of existing solutions: it is multi-scale, meaning it can identify and uniformly anonymize objects of different scales in the same image, and it is class-generic, so it can be used to anonymize any class …

Data Protection In Nigeria: Addressing The Multifarious Challenges Of A Deficient Legal System, Roland Akindele

Journal of International Technology and Information Management

This paper provides an overview of the current state of privacy and data protection policies and regulations in Nigeria. The paper contends that the extant legal regime in Nigeria is patently inadequate to effectively protect individuals against abuse resulting from the processing of their personal data. The view is based on the critical analysis of the current legal regime in Nigeria vis-à-vis the review of some vital data privacy issues. The paper makes some recommendations for the reform of the law.

Trust And Persistence, Paul A. Buhler, Michael N. Huhns

Faculty Publications

We rely on computers to control our power plants and water supplies, our automobiles and transportation systems, and soon our economic and political systems. Increasingly, software agents are enmeshed in these systems, serving as the glue that connects distributed components. Clearly, we need mechanisms to determine whether these agents are trustworthy. What do we need to establish trust? Agents are often characterized by features such as autonomy, sociability, proactiveness, and persistent identity. This latter feature is key in determining trust. When agents operate over an extended period, they can earn a reputation for competence, timeliness, ease of use, and trustworthiness, …