Use of Machine Learning (ML) to understand underlying patterns in gene mutations (genomics) has far-reaching results in diagnosis and treatment for life-threatening diseases like cancer. Success and sustainability of ML algorithms depends on the quality and diversity of training data, and under-representation of groups (gender, race, etc.) can lead to exacerbation of systemic discrimination issues. In this work, we propose TRAPDOOR, a methodology for the identification of biased datasets by repurposing, otherwise malicious, neural backdoors. Our methodology can leak potential bias information about the cloud’s dataset which is collected in a collaborative setting, without hampering the genuine performance. Using a real-world cancer genomics dataset, we analyze feasibility of leaking bias for gender and race attributes. Our experimental results show that TRAPDOOR can detect the presence of dataset bias with 100% accuracy, and furthermore can also extract the extent of bias by recovering the percentage with a small error.

Digitization has been rapidly integrated with manufacturing industries and critical infrastructure to increase efficiency, productivity, and reduce wastefulness, a transition being labeled as Industry 4.0. However, this expansion, coupled with the poor cybersecurity posture of these Industrial Internet of Things (IIoT) devices, has made them prolific targets for exploitation. Moreover, modern Programmable Logic Controllers (PLC) used in the Operational Technology (OT) sector are adopting open-source operating systems such as Linux instead of proprietary software, making such devices susceptible to Linux-based malware. Traditional malware detection approaches cannot be applied directly or extended to such environments due to the unique restrictions of these PLC devices, such as limited computational power and real-time requirements. In this paper, we propose ORRIS, a novel lightweight and out-of-the-device framework that detects malware at both kernel and user-level by processing the information collected using the Joint Test Action Group (JTAG) interface. We evaluate ORRIS against in-the-wild Linux malware achieving maximum detection accuracy of ≈99.7% with very few false-positive occurrences, a result comparable to the state-of-the-art commercial products. Moreover, we also develop and demonstrate a real-time implementation of ORRIS for commercial PLCs.

Recent work has shown that the introduction of autonomous vehicles (AVs) in traffic could help reduce traffic jams. Deep reinforcement learning methods demonstrate good performance in complex control problems, including autonomous vehicle control, and have been used in state-of-the-art AV controllers. However, deep neural networks (DNNs) render automated driving vulnerable to machine learning-based attacks. In this work, we explore the backdooring/trojanning of DRL-based AV controllers. We develop a trigger design methodology that is based on well-established principles of traffic physics. The malicious actions include vehicle deceleration and acceleration to cause stop-and-go traffic waves to emerge (congestion attacks) or AV acceleration resulting in the AV crashing into the vehicle in front (insurance attack). We test our attack on single-lane and two-lane circuits. Our experimental results show that the backdoored model does not compromise normal operation performance, with the maximum decrease in cumulative rewards being 1%. Still, it can be maliciously activated to cause a crash or congestion when the corresponding triggers appear.

Malicious manipulations on Industrial Control Systems (ICSs) endanger critical infrastructures, causing unprecedented losses. State-of-the-art research in the discovery and exploitation of vulnerability typically assumes full visibility and control of the industrial process, which in real-world scenarios is unrealistic. In this work, we investigate the possibility of an automated end-to-end attack for an unknown control process in the constrained scenario of infecting just one industrial computer. We create databases of human-machine interface images, and Programmable Logic Controller (PLC) binaries using publicly available resources to train machine-learning models for modular and granular fingerprinting of the ICS sectors and the processes, respectively. We then explore control-theoretic attacks on the process leveraging common/ubiquitous control algorithm modules like Proportional Integral Derivative blocks using a PLC binary reverse-engineering tool, causing stable or oscillatory deviations within the operational limits of the plant. We package the automated attack and evaluate it against a benchmark chemical process, demonstrating the feasibility of advanced attacks even in constrained scenarios.