In autonomous and self-driving vehicles, visual perception of the driving environment plays a key role. Vehicles rely on machine learning (ML) techniques such as deep neural networks (DNNs), which are extensively trained on manually annotated databases to achieve this goal. However, the availability of training data that can represent different environmental conditions can be limited. Furthermore, as different driving terrains require different decisions by the driver, it is tedious and impractical to design a database with all possible scenarios. This work proposes a semi-parametric approach that bypasses the manual annotation required to train vehicle perception systems in autonomous and self-driving vehicles. We present a novel “Performance and Environment-aware Advanced Driving Assistance Systems” which employs one-shot learning for efficient data generation using user action and response in addition to the synthetic traffic data generated as Pareto optimal solutions from one-shot objects using a set of generalization functions. Adapting to the driving environments through such optimization adds more robustness and safety features to autonomous driving. We evaluate the proposed framework on environment perception challenges encountered in autonomous driving assistance systems. To accelerate the learning and adapt in real-time to perceived data, a novel deep learning-based Alternating Direction Method of Multipliers (dlADMM) algorithm is introduced to improve the convergence capabilities of regular machine learning models. This methodology optimizes the training process and makes applying the machine learning model to real-world problems more feasible. We evaluated the proposed technique on AlexNet and MobileNetv2 networks and achieved more than 18$\times$speedup. By making the proposed technique behavior-aware we observed performance of upto 99% while detecting traffic signals.

Networked IoT systems have emerged in recent years to facilitate seamless connectivity, portability, and smarter functionality. Despite lending a plethora of benefits, IoT devices are exploited by adversaries for various illicit purposes. IoT systems are a popular target due to the lack of security traits in the design, and minimal available computational and storage resources on the devices. Among multiple threats, malicious applications a.k.a malware are seen as a pivotal security threat on IoT devices and networks. Many malware detection techniques have been proposed recently. However, the existing techniques focus either on general-purpose systems or assume the availability of abundant resources at their disposal for malware detection. However, for IoT devices, the ongoing workloads such as sensing, and on-device computations further minimize the available resources for malware detection. We propose a novel resource- and workload-aware malware detection integrated with distributed computing for IoT networked systems to address these challenges. The device analyzes the available resources for malware detection using a lightweight regression model. Depending on the available resources, ongoing workload executions, and communication cost the malware detection task is either performed on-device or offloaded to neighboring IoT nodes with sufficient resources. To ensure data integrity and user privacy, instead of offloading the whole malware detection, the classifier is partitioned and distributed over multiple nodes and further integrated at the parent node for malware detection. Experimental analysis shows that the proposed technique can achieve a speed-up of $9.8 \times$ compared to on-device inference while maintaining a malware detection accuracy of 96.7%.

Cybersecurity in recent years has gained significant attention, especially with the deployment of millions of devices across the globe and increased threats targeted toward embedded systems. Many cyber threats have been detected and emerged in the last few years. Among multiple threats, malware attacks are considered to be prominent due to the impact on users and systems. Considering the evolving trend of such cyber threats, traditional statistical and heuristic threat detection approaches have observed the need to be more effective and efficient. Machine learning (ML)-based cyber-threat detection has been actively researched and adopted across academia and industry to address the challenges of evolving cyber threats. However, ML-based neural network techniques though efficient, are considered black boxes due to the lack of sufficient information that can be used to deduce their functionality. On the other hand, the interpretable and explainable AI/ML field focuses on the explainability and reason for the decisions performed by the ML models. In this paper, we experiment with different explainable AI (XAI) techniques for interpreting multiple malware detection models. Specifically, we analyze the consistency and reliability of these neural network models in determining an attack and benign functions. We provide quantitative analysis of multiple explanation methods across different datasets. When trained with the top feature attributes (10%-35% of whole data) generated by XAI methods, the ML classifiers (trained on High Performance Counters and Mimicus PDF malware datasets) retain a malware detection accuracy of 88%-92%. The ML classifiers are also compared with state-of-the-art models and the proposed technique (training with partial data features generated by explainable methods) produce comparable malware detection accuracy above 82%.

The wide adoption of Internet of Things (IoT) devices has led to better connectivity along with seamless communication and smart computation capabilities across the network. Despite being deployed widely across the globe, IoT devices are prominently exploited for security vulnerabilities due to the lack of inherent security measures. Among multiple threats, malicious applications also known as malware is a pivotal security threat for IoT devices. Lack of security traits and limited resources are the primary hindrances for the adoption of existing malware detection techniques in IoT devices. Furthermore, the existing techniques assume the availability of all the device resources for malware detection. However, for IoT devices deployed for critical real-world applications, the available on-device resources for a given task, including malware detection are minimal compared to the overall available resources. To address this primary challenge, this work introduces a novel resource- and workload-aware model-parallelism-inspired malware detection for IoT devices. The device first analyzes the available resources for malware detection using a lightweight regression model. Depending on the available resources, ongoing workload executions, and communication costs, the malware detection task is either performed on-device or offloaded to neighboring IoT nodes with sufficient resources. To ensure data integrity and user privacy, instead of offloading the whole malware detection, the classifier is partitioned and distributed over multiple nodes and further integrated at the parent node for malware detection. Experimental analysis shows that the proposed technique can achieve a speed-up of$9.8\times $compared to on-device inference while maintaining a malware detection accuracy of 96.7%.

One of the pivotal security threats for embedded computing systems is malicious softwarea.k.a malware. With efficiency and efficacy, Machine Learning (ML) has been widely adopted for malware detection in recent times. Despite being efficient, the existing techniques require updating the ML model frequently with newer benign and malware samples for training and modeling an efficient malware detector. Furthermore, such constraints limit the detection of emerging malware samples due to the lack of sufficient malware samples required for efficient training. To address such concerns, we introduce a code-aware data generation-based few-shot learning technique. CAD-FSL generates multiple mutated samples of the limitedly seen malware for efficient malware detection. Loss minimization ensures that the generated samples closely mimic the limitedly seen malware, restore malware functionality and mitigate the impractical samples. Such developed synthetic malware is incorporated into the training set to formulate the model that can efficiently detect the emerging malware despite having limited (few-shot) exposure. The experimental results demonstrate that with the proposed "Code-Aware Data Generation" technique, we detect malware with 90% accuracy, which is approximately 9% higher while training classifiers with only limitedly available training data.