FedDroidMeter: A Privacy Risk Evaluator for FL-Based Android Malware Classification Systems.

In traditional centralized Android malware classifiers based on machine learning, the training sample uploaded by users contains sensitive personal information, such as app usage and device security status, which will undermine personal privacy if used directly by the server. Federated-learning-based Android malware classifiers have attracted much attention due to their privacy-preserving and multi-party joint modeling. However, research shows that indirect privacy inferences from curious central servers threaten this framework. We propose a privacy risk evaluation framework, FedDroidMeter, based on normalized mutual information in response to user privacy requirements to measure the privacy risk in FL-based malware classifiers. It captures the essential cause of the disclosure of sensitive information in classifiers, independent of the attack model and capability. We performed numerical assessments using the Androzoo dataset, the baseline FL-based classifiers, the privacy-inferred attack model, and the baseline methodology of privacy evaluation. The experimental results show that FedDroidMeter can measure the privacy risks of the classifiers more effectively. Meanwhile, by comparing different models, FL, and privacy parameter settings, we proved that FedDroidMeter could compare the privacy risk between different use cases equally. Finally, we preliminarily study the law of privacy risk in classifiers. The experimental results emphasize the importance of providing a systematic privacy risk evaluation framework for FL-based malware classifiers and provide experience and a theoretical basis for studying targeted defense methods.

High-Fat Diet Consumption in Adolescence Induces Emotional Behavior Alterations and Hippocampal Neurogenesis Deficits Accompanied by Excessive Microglial Activation.

Adolescence is a developmental epoch characterized by massive neural circuit remodeling; thus, the brain is particularly vulnerable to environmental influences during this period. Excessive high-fat diet (HFD) consumption, which is very common among adolescents, has long been recognized as a potent risk factor for multiple mood disorders, including depression and anxiety. However, the precise mechanisms underlying the influences of HFD consumption in adolescence on emotional health are far from clear. In the present study, C57BL/6 mice were fed a control diet (CD) or HFD for about 4 weeks from postnatal day (P) 28 to P60, spanning most of the adolescence period, and then subjected to behavioral assessments and histological examinations. HFD mice exhibited elevated levels of depression and anxiety, decreased hippocampal neurogenesis, and excessive microglial activation in the ventral hippocampus. Furthermore, in HFD-fed mice, microglia showed increased DCX+ inclusions, suggesting aberrant microglial engulfment of newborn neurons in HFD-fed adolescents. To our knowledge, this is the first observation suggesting that the negative effects of HFD consumption in adolescence on emotion and neuroplasticity may be attributed at least in part to aberrant microglial engulfment of nascent neurons, extending our understanding of the mechanism underlying HFD-related affective disorders in young people.