Traditionally, authentication systems have followed a non-feedback approach, requiring users to present credentials before accessing resources. Over time, users have become accustomed to this oblivious form of authentication. However, this model offers no opportunity for users to provide feedback that could enhance the system's effectiveness. Similarly, biometric-based implicit authentication, while transparent, often excludes users entirely from the feedback loop. Incorporating user feedback into authentication systems has the potential to significantly improve their performance. Achieving this, however, requires a novel framework capable of standardizing user input, extracting meaningful information from feedback, and integrating the user more closely into the system. To this end, we challenge conventional authentication paradigms and introduce User-Centered Implicit Authentication (UserIA)-a customizable approach that extends beyond the limits of traditional schemes. UserIA maintains the transparency of implicit authentication while delivering improved accuracy and reduced overhead. To enable secure feedback-driven feature adaptation, UserIA introduces a new technique calledbehavior alignment. Additionally, it appliesuser operant conditioningfrom psychology to reinforce user behavior and further enhance authentication accuracy. We have implemented and thoroughly evaluated UserIA in a real-world environment. Experimental results demonstrate that UserIA achieves a lower Equal Error Rate (EER) and consumes less time and energy compared to existing methods.

In the advancing field of machine learning, balancing accuracy, interpretability, and privacy represents a significant challenge. The problem is exacerbated by the widespread deployment of pre-trained models locally in diverse applications, which could lead to various amounts of privacy leakage. Conventional Differential Privacy strategies, in which uniform noises are applied to model gradients, guarantee data privacy at the expense of accuracy and interpretability. This paper introduces a Feature-Sensitive Adaptive Differential Privacy (FADP) framework with a unique noise-adding strategy. Noises are adaptively added based on feature importance clustering, where important features are considered for interpretability. By employing a unique masking technique, FADP selectively preserves crucial features with minimal noise interference, maintaining accuracy while enhancing interpretability. The FADP framework addresses the limitations of traditional DP methods by preserving critical channels and improving interpretability — a vital requirement in machine learning applications that demand transparency in model decisions. Through comprehensive testing, FADP is shown to balance the trade-offs among accuracy, privacy, and interpretability, marking a substantial advancement in the field of privacy-preserving machine learning.