This paper investigates the security vulnerabilities of prompt-learning-based FL systems in a healthcare setting. Specifically, we use a backdoor attack that leverages learnable prompt vectors in vision-language medical foundation models to execute stealthy adversarial manipulations. We evaluate our attack across diverse healthcare datasets and FL configurations, showing that while FL is useful as a privacy-preserving mechanism, it is susceptible to targeted backdoor attacks that pose a threat to medical applications.

Autonomous robots, autonomous vehicles, and humans wearing mixed-reality headsets require accurate and reliable tracking services for safety-critical applications in dynamically changing real-world environments. However, the existing tracking approaches, such as Simultaneous Localization and Mapping (SLAM), do not adapt well to environmental changes and boundary conditions despite extensive manual tuning. On the other hand, while deep learning-based approaches can better adapt to environmental changes, they typically demand substantial data for training and often lack flexibility in adapting to new domains. To solve this problem, we propose leveraging the neurosymbolic program synthesis approach to construct adaptable SLAM pipelines that integrate the domain knowledge from traditional SLAM approaches while leveraging data to learn complex relationships. While the approach can synthesize end-to-end SLAM pipelines, we focus on synthesizing the feature extraction module. We first devise a domain-specific language (DSL) that can encapsulate domain knowledge on the essential attributes for feature extraction and the real-world performance of various feature extractors. Our neurosymbolic architecture then undertakes adaptive feature extraction, optimizing parameters via learning while employing symbolic reasoning to select the most suitable feature extractor. Our evaluations demonstrate that our approach, neurosymbolic Feature EXtraction (nFEX), yields higher-quality features. It also reduces the pose error observed for the state-of-the-art baseline feature extractors ORB and SIFT by up to 90% and up to 66%, respectively, thereby enhancing the system’s efficiency and adaptability to novel environments.

Data heterogeneity among Federated Learning (FL) users poses a significant challenge, resulting in reduced global model performance. The community has designed various techniques to tackle this issue, among which Knowledge Distillation (KD)-based techniques are common. While these techniques effectively improve performance under high heterogeneity, they inadvertently cause higher accuracy degradation under model poisoning attacks (known as \emph{attack amplification}). This paper presents a case study to reveal this critical vulnerability in KD-based FL systems. We show why KD causes this issue through empirical evidence and use it as motivation to design a hybrid distillation technique. We introduce a novel algorithm, Hybrid Knowledge Distillation for Robust and Accurate FL (HYDRA-FL), which reduces the impact of attacks in attack scenarios by offloading some of the KD loss to a shallow layer via an auxiliary classifier. We model HYDRA-FL as a generic framework and adapt it to two KD-based FL algorithms, FedNTD and MOON. Using these two as case studies, we demonstrate that our technique outperforms baselines in attack settings while maintaining comparable performance in benign settings.

Universal time is critical for coordinating function-alities among co-located as well as geographically distributed IoT devices. Current time alignment approaches for IoT devices rely on radio-based communication that puts an extra burden on the already resource-constrained devices. Other timing approaches depend either on customized hardware frontends or on fixed networking capabilities. Intermittent network connectivity fur-ther deteriorates timing performance for these devices. These constraints motivate us to create a new design that actively embeds time information into the surroundings of IoT devices that can harvest timing signals with off-the-shelf sensing capabil-ities. Our design decouples clock performance from the network uncertainties, introduce resource efficiency and extensibility, and requires no modification in existing devices. A unique property of our design is that it leverages the ubiquitous Electric Network Frequency (ENF) fluctuations as global time reference for the sensing devices, and takes on a variety of challenges at the in-tersection of sensing and signal processing to provide a universal sense of time without custom hardware frontends and network dependability. We evaluate the extensibility of our design in remote setups and show its robustness in real world settings.

Split Learning (SL) and Federated Learning (FL) are two prominent distributed collaborative learning techniques that maintain data privacy by allowing clients to never share their private data with other clients and servers, and find extensive IoT applications in smart healthcare, smart cities and smart industry. Prior work has extensively explored the security vulnerabilities of FL in the form of poisoning attacks. To mitigate the effect of these attacks, several defenses have also been proposed. Recently, a hybrid of both learning techniques has emerged (commonly known as SplitFed) that capitalizes on their advantages (fast training) and eliminates their intrinsic disadvantages (centralized model updates).