Federated Learning (FL) is a privacy-preserving approach to distributed machine learning, but it is vulnerable to inference attacks. Inference attacks aim to extract information about regarding model or data used in FL, directly threatening one of its core principles. This essay provides an overview of state-of-the-art inference attacks in FL and their implications for data privacy. It introduces the basics of FL, different types of inference attacks (model inversion, membership inference, property inference), and provides an overview of recent research on gradient inversion and membership inference attacks. We emphasize the need for robust defenses to safeguard sensitive information in FL systems, along with the importance of future research in addressing these increasingly realistic threats.

Figure

Listing

# List of Figures


# List of Listings


# List of Tables


Table

Inference Attacks on Federated Learning - A Survey

CTF Challenge VulnLawyers Writeup

The field of automated negotiation promises to improve negotiations, thus, a fair outcome and process should also be considered when building these systems. However, issues exist with computational approaches to fairness with which the field of computer science is mainly concerned. To this end, we propose a new approach to fairness based on that of essentially contested concepts to see if argumentation-based negotiation could be used as an extension to the Stacked Alternating Offers Protocol to improve fairness. Looking at fairness as an essentially contested concept shows that discussion between people somehow influenced by the negotiation system is necessary to maintain its fairness. This in turn means that systems that provide accessible context are fairer than systems that would not do so. Thus arguments, if implemented in an accessible manner, add more context to the negotiation, in turn making an SAOP negotiation fairer.

A non-computational approach to Fairness in the context of negotiations. My Bachelor’s thesis.

All Posts

Posts