Penetration testing, or pentesting, refers to assessing network system security by trying to identify and exploit any existing vulnerabilities. Reinforcement Learning (RL) has recently become an effective method for creating autonomous pentesting agents. However, RL agents are typically trained in a simulated network environment. This can be challenging when deploying them in a real network infrastructure due to the lack of realism of the simulation-trained agents. In this paper, we present PenGym, a framework for training pentesting RL agents in realistic network environments. The most significant features of PenGym are its support for real pentesting actions, full automation of the network environment creation, and good execution performance. The results of our experiments demonstrated the advantages and effectiveness of using PenGym as a realistic training environment in comparison with a simulation approach (NASim). For the largest scenario, agents trained in the original NASim environment behaved poorly when tested in a real environment, having a high failure rate. In contrast, agents trained in PenGym successfully reached the pentesting goal in all our trials. Even after fixing logical modeling issues in simulation to create the revised version NASim(rev.), experiment results with the largest scenario indicated that agents trained in PenGym slightly outperformed, and were more stable, than those trained in NASim(rev.). Thus, the average number of steps required to reach the pentesting goal was 1.4 to 8 steps better for PenGym. Consequently, PenGym provides a reliable and realistic training environment for pentesting RL agents, eliminating the need to model agent actions via simulation.

In the fourth industrial revolution, securing the protection of supply chains has become an ever-growing concern. One such cyber threat is a hardware Trojan (HT), a malicious modification to an IC. HTs are often identified during the hardware manufacturing process but should be removed earlier in the design process. Machine learning-based HT detection in gate-level netlists is an efficient approach to identifying HTs at the early stage. However, feature-based modeling has limitations in terms of discovering an appropriate set of HT features. We thus proposeNHTD-GLin this paper, a novel node-wise HT detection method based on graph learning (GL). Given the formal analysis of the HT features obtained from domain knowledge,NHTD-GLbridges the gap between graph representation learning and feature-based HT detection. The experimental results demonstrate thatNHTD-GLachieves 0.998 detection accuracy and 0.921 F1-score and outperforms state-of-the-art node-wise HT detection methods.NHTD-GLextracts HT features without heuristic feature engineering.

As security risks to network systems have grown, red team assessment has emerged as a powerful methodology for discovering vulnerabilities.Such assessments are difficult to master because technical knowledge and experience are needed.Automating the vulnerability assessment of network systems is expected to help network system administrators conduct these assessments easily.The challenges for automating these assessments include accurately addressing many actions, observing network states, and generalizing agent models.In this paper, we propose a framework, called AutoRed, for the automation of red team assessment via strategic thinking using reinforcement learning (RL).Our framework addresses the following challenges: (1) facilitating action determination by adopting a hierarchical RL model via strategic thinking, (2) establishing a method to observe network systems using graph neural networks (GNNs), and (3) investigating the reusability and generalization ability of the proposed model through experiments.We further evaluate the proposed model in an emulated environment constructed on a virtual machine platform.The experimental results demonstrate that the proposed model trained on three scenarios simultaneously can be applied 10-40 times more efficiently to various scenarios, including unseen scenarios during training, than the state-of-the-art hierarchical model.

Large language models (LLMs) have recently been employed for a wide variety of purposes. Retrieval-augmented generation (RAG), in which an LLM generates a response based on context relevant to the prompt, is often used to enable the LLM to adapt to specialized domains. However, sentences generated by a generative LLM may contain incorrect information, known as “hallucinations.” The challenge in identifying hallucinations within the RAG framework involves evaluating the certainty of both context retrieval and LLM outputs. In this paper, we propose a metric called RAG certainty to quantify the certainty of LLM outputs within a RAG framework. The proposed metric is calculated based on certainty scores from both information retrieval and response generation. Experimental results demonstrate that the proposed metric effectively reflects the certainty of information retrieval in a RAG framework. We further validated the proposed metric through a case study that assesses the predicted Common Vulnerability Scoring Sys-tem (CVSS) scores for cybersecurity vulnerabilities and found that errors are mitigated according to the proposed metric.

Graph neural networks (GNNs) have been actively employed in hardware security and have demonstrated remarkable performance. In particular, GNN models for hardware Trojan (HT) detection significantly outperform existing machine learning-based detection methods. However, GNNs have a potential vulnerability to membership inference attack (MIA), which aims to determine whether a given sample is used in the training dataset. In this paper, we investigate the threat of MIAs for GNN-based HT detection models. First, the MIA scheme for GNN-based HT detection models is established based on the basic MIA settings. The experimental results demonstrate that MIA for GNN-based HT detection can leak information about the HTs included in the training dataset with a 0.945 attack AUC score in the worst-case scenario. Based on this observation, we propose a defense method against MIA utilizing a domain generalization technique. The proposed defense method successfully mitigated the vulnerability of MIA and degraded the attack AUC score to 0.536 for the netlist level while maintaining the original HT detection performance.

Hardware Trojans (HTs) have become a serious problem, and extermination of them is strongly required for enhancing the security and safety of integrated circuits. An effective solution is to identify HTs at the gate level via machine learning techniques. However, machine learning has specific vulnerabilities, such asadversarial examples. In reality, it has been reported that adversarial modified HTs greatly degrade the performance of a machine learning-based HT detection method. Therefore, we propose a robust HT detection method using adversarial training (R-HTDetector). We formally describe the robustness of R-HTDetector in modifying HTs. Our work gives the world-first adversarial training for HT detection with theoretical backgrounds. We show through experiments with Trust-HUB benchmarks that R-HTDetector overcomes adversarial examples while maintaining its original accuracy.

Recently, with the increase in outsourcing of IC design and manufacturing, the possibility of inserting hardware Trojans, which are circuits with malicious functions, has been pointed out. To prevent this threat, a method to identify hardware Trojans using neural networks has been proposed. On the other hand, adversarial attacks have emerged that modify circuit design information to reduce the accuracy of hardware-Trojan classification by neural networks. Since the features designed by existing methods do not take the attacks into account, it is necessary to consider a new method for countermeasures. In this paper, out of 76 features that are strongly related to hardware-Trojan features, we investigate them from the viewpoint of the robustness against the adversarial attacks on circuit design information and newly propose 24 hardware-Trojan features. We compare the classifiers using the proposed 24 features with the classifiers using 11, 36, 51, and 76 existing features, respectively and confirm that the proposed ones are more robust in identifying hardware Trojans in circuits subjected to the adversarial attacks.

Due to the rapid growth in the information and telecommunications industries, an untrusted vendor might compromise the complicated supply chain by inserting hardware Trojans (HTs). Although hardware Trojan detection methods at gate-level netlists employing machine learning have been developed, the training dataset is insufficient. In this paper, we propose a data augmentation method for machine-learning-based hardware Trojan detection. Our proposed method replaces a gate in a hardware Trojan circuit with logically equivalent gates. The experimental results demonstrate that our proposed method successfully enhances the classification performance with all the classifiers in terms of the true positive rates (TPRs).

A heuristic algorithm is one of the approaches to solve an NP-hard problem. In order to enhance the capability of the system, heterogeneous computing is often adapted. In this paper, we propose an FPGA-based heterogeneous solver for three-dimensional routing. The proposed system is implemented into multiple FPGA boards and a single-board computer. The experimental results demonstrate that the proposed system outperforms a single FPGA system.

Recently, with the spread of Internet of Things (IoT) devices, embedded hardware devices have been used in a variety of everyday electrical items. Due to the increased demand for embedded hardware devices, some of the IC design and manufacturing steps have been outsourced to third-party vendors. Since malicious third-party vendors may insert hardware Trojans into their products, developing an effective hardware Trojan detection method is strongly required. In this paper, we evaluate hardware Trojan detection methods using neural networks and random forests at gate-level intellectual property (IP) cores that contain more than 10,000 nets. First, we extract 11 features for each net in a given netlist, and learn them with neural networks and random forests. Then, we classify the nets in an unknown netlist into a set of normal nets and Trojan nets based on the learned classifiers. The experimental results demonstrate that the average true positive rate becomes 84.6% and the average true negative rate becomes 95.1%, which is sufficiently high accuracy compared to existing evaluations.

With the widespread use of Internet of Things (IoT) devices in recent years, we utilize a variety of hardware devices in our daily life. On the other hand, hardware security issues are emerging. Power analysis is one of the methods to detect anomalous operations, but it is hard to apply it to IoT devices where an operating system and various software programs are running. In this paper, we propose an anomalous behavior detection method for an IoT device by extracting application-specific power behaviors. First, we measure a power consumption of an IoT device, and obtain the power waveform. Next, we extract an application-specific power waveform by eliminating a steady factor from the obtained power waveform. Finally, we extract feature values from the application-specific power waveform and detect an anomalous behavior by utilizing the local outlier factor (LOF) method. The experimental results using a single board computer demonstrate that the proposed method successfully detects the anomalous power behavior of an anomalous application program.

Since hardware/software vendors produce their IoT products easily and inexpensively, they often outsource their designs to third-party vendors where malicious third-party vendors can have a chance to insert software Trojans as well as “hardware Trojans” into their IoT devices. How to tackle the issue becomes a serious concern these days. In this paper, we propose an anomaly behavior detection method utilizing accurate power analysis for low-cost micro-controllers. Our method accurately measures power consumption of the target device, and then classifies its waveform into the sleep-mode part, in which a micro-controller saves power, and into the active-mode part, in which a micro-controller works in a normal operation. After that, we obtain the duration time and consumed power from each active-mode period as feature values. Finally, we detect abnormal behavior based on the obtained feature values utilizing an outlier detection method. In our experiments, we empirically evaluate the proposed method utilizing two types of micro-controllers, and the experimental results demonstrate that our proposed method successfully detects abnormal behaviors.

Microcontrollers are widely used in electric devices such as smart phones, televisions, and other smart IoT (Internet-of-Things) devices. Because of the increase of these smart IoT devices, the security of hardware devices becomes a serious concern. In this paper, we propose a method which detects the existence of malfunctions implemented in microcontrollers utilizing power analysis. Our method firstly measures power consumption of the target device and classifies its waveform into the sleep-mode part, in which a microcontroller saves power, and the active-mode part, in which a microcontroller works in a normal operation. After that, we focus on the active-mode part and extract several features from the waveform, which effectively distinguish between normal operations and malfunctions. Finally, we classify the features and identify whether malfunctions exist or not. Our experimental results demonstrate that our proposed method successfully detects the existence of malfunctions in our benchmark.

Recently, high-functioning hardware devices such as smart TVs and smart phones have been widely used in our daily lives. To keep up with the rapid advance of these high technologies, reconfigurable hardware devices such as FP-GAs (Field Programmable Gate Arrays) have been used in final products. Under the circumstances, the risks that mal-functions may be inserted into hardware devices have arisen. The malfunctions inserted into hardware devices are known as hardware Trojans. How to detect them becomes serious concern in hardware production. In this paper, we design a Trojan-infected cryptographic circuit as well as a Trojan-invalidating circuit, and implement them on an FPGA board. To begin with, we design an AES cryptographic circuit. Secondly, we insert a hardware Trojan into the AES cryptographic circuit. Finally, we design a Trojan-invalidating circuit and insert it into a suspicious Trojan net in the Trojan-infected cryptographic circuit. After that, we implement the circuits into an FPGA board. The experimental results demonstrate that the Trojan-invalidating circuit adequately deactivate the suspicious Trojan net in the Trojan-infected cryptographic circuit.

Recently, due to the increase of outsourcing in IC design and manufacturing, it has been reported that malicious third-party IC vendors often insert hardware Trojans into their products. Especially in IC design step, it is strongly required to detect hardware Trojans because malicious third-party vendors can easily insert hardware Trojans in their products. In this paper, we propose a machine-learning-based hardware-Trojan detection method for gate-level netlists using multi-layer neural networks. First, we extract 11 Trojan-net feature values for each net in a netlist. After that, we classify the nets in an unknown netlist into a set of Trojan nets and that of normal nets using multi-layer neural networks. We obtained at most 100% true positive rate with our proposed method.

Recently, due to the increase of outsourcing in IC design, it has been reported that malicious third-party vendors often insert hardware Trojans into their ICs. How to detect them is a strong concern in IC design process. The features of hardware-Trojan infected nets (or Trojan nets) in ICs often differ from those of normal nets. To classify all the nets in netlists designed by third-party vendors into Trojan ones and normal ones, we have to extract effective Trojan features from Trojan nets. In this paper, we first propose 51 Trojan features which describe Trojan nets from netlists. Based on the importance values obtained from the random forest classifier, we extract the best set of 11 Trojan features out of the 51 features which can effectively detect Trojan nets, maximizing the F-measures. By using the 11 Trojan features extracted, the machine-learning based hardware Trojan classifier has achieved at most 100% true positive rate as well as 100% true negative rate in several TrustHUB benchmarks and obtained the average F-measure of 74.6%, which realizes the best values among existing machine-learning-based hardware-Trojan detection methods.

Recently, we face a serious risk that malicious third-party vendors can very easily insert hardware Trojans into their IC products but it is very difficult to analyze huge and complex ICs. In this paper, we propose a hardware-Trojan classification method to identify hardware-Trojan infected nets (or Trojan nets) using a support vector machine (SVM). Firstly, we extract the five hardware-Trojan features in each net in a netlist. Secondly, since we cannot effectively give the simple and fixed threshold values to them to detect hardware Trojans, we represent them to be a five-dimensional vector and learn them by using SVM. Finally, we can successfully classify a set of all the nets in an unknown netlist into Trojan ones and normal ones based on the learned SVM classifier. We have applied our SVM-based hardware-Trojan classification method to Trust-HUB benchmarks and the results demonstrate that our method can much increase the true positive rate compared to the existing state-of-the-art results in most of the cases. In some cases, our method can achieve the true positive rate of 100%, which shows that all the Trojan nets in a netlist are completely detected by our method.