Epidemics pose a significant threat to societal development. Accurately and swiftly identifying the source of an outbreak is crucial for controlling the spread of an epidemic and minimizing its impact. However, existing research on locating epidemic sources often overlooks the fact that epidemics have an incubation period and fails to consider social behaviors like self-isolation during the spread of the epidemic. In this study, we first take into account isolation behavior and introduce the Susceptible-Exposed-Infected-Recovered (SEIR) propagation model to simulate the spread of epidemics. As the epidemic reaches a certain threshold, government agencies or hospitals will report the IDs of some infected individuals and the time when symptoms first appear. The reported individuals, along with their first and second-order neighbors, are then isolated. Using the moment of symptom onset reported by the isolated individuals, we propose a node-level classification method and subsequently develop the node-level-based source identification (NLSI) algorithm. Extensive experiments demonstrate that the NLSI algorithm is capable of solving the source identification problem for single and multiple sources under the SEIR propagation model. We find that the source identification accuracy is higher when the infection rate is lower, and a sparse network structure is beneficial to source localization. Furthermore, we discover that the length of the isolation period has little impact on source localization, while the length of the incubation period significantly affects the accuracy of source localization. This research offers a novel approach for identifying the origin of the epidemic associated with our defined SEIR model.

Accurately locating sources with limited resources plays an important role in suppressing and even predicting the spread of harmful information. Due to the existence of the coupling between layers, compared with simplex networks, many studies have revealed that multiplex networks present many interesting phenomena. However, almost all the works involved in source(s) localization concentrate on simplex networks, while it is unclear how the coupling strength between layers of a multiplex network affects source(s) localization. Here we focus on a linear diffusion process and propose a general framework of multiple sources localization in multiplex networks with minimum measurements by combining the observability theory and compressed sensing theory. One surprise finding is that, for undirected two-layer networks with fully coupling between layers, the messengers placed at the same layer is sufficient to accurately locate all the sources. Different from simplex networks, the diffusion coefficients of linear diffusion process in multiplex networks can affect the minimum number of messengers. Simulation results from various network structures show that the minimum number of messengers is fewer when the coupling strength between layers is higher. Besides, we find the larger the network size the fewer measurements are required for sources localization. We further provide simulations from synthetic networks and one real network demonstrates our framework of sources localization is robust against small amount of data, noise and the coupling strength between layers.

Existing research shows that cooperative multi-agent deep reinforcement learning (c-MADRL) is vulnerable to adversarial attacks, and c-MADRL is increasingly being applied to safety-critical domains. However, the robustness of c-MADRL against adversarial attacks has not been fully studied. In the setting of c-MADRL, unlike the single-agent scenario, an adversary can attack multiple agents or all agents at each time step, but the attacker needs more computation to generate adversarial examples and will be more easily detected. Therefore, how the attacker chooses one or several agents instead of all agents to attack is a significant issue in the setting of c-MADRL. Aiming to address this issue, this paper proposes a novel adversarial attack approach, which dynamically groups the agents according to relevant features and selects a group to attack based on the group’s contribution to the overall reward, thus effectively reducing the cost and number of attacks, as well as improving attack efficiency and decreasing the chance of attackers being detected. Moreover, we exploit the transferability of adversarial examples to greatly reduce the computational cost of generating adversarial examples. Our method is tested in multi-agent particle environments (MPE) and in StarCraft II. Experimental results demonstrate that our proposed method can effectively degrade the performance of multi-agent deep reinforcement learning algorithms with fewer attacks and lower computational costs.

Human activity recognition(HAR) from the temporal streams of sensory data has been applied to many fields, such as healthcare services, intelligent environments and cyber security. However, the classification accuracy of most existed methods is not enough in some applications, especially for healthcare services. In order to improving accuracy, it is necessary to develop a novel method which will take full account of the intrinsic sequential characteristics for time-series sensory data. Moreover, each human activity may has correlated feature relationship at different levels. Therefore, in this paper, we propose a three-stage continuous hidden Markov model (TSCHMM) approach to recognize human activities. The proposed method contains coarse, fine and accurate classification. The feature reduction is an important step in classification processing. In this paper, sparse locality preserving projections (SpLPP) is exploited to determine the optimal feature subsets for accurate classification of the stationary-activity data. It can extract more discriminative activities features from the sensor data compared with locality preserving projections. Furthermore, all of the gyro-based features are used for accurate classification of the moving data. Compared with other methods, our method uses significantly less number of features, and the over-all accuracy has been obviously improved.

Although the deep reinforcement learning (DRL) technology has been widely adopted in various fields, it has become an important research hotspot to study the vulnerability of DRL for improving the robustness of DRL agents. The adversarial attack methods based on white-box models, where the adversary can access all the information of victims, have been intensively investigated. However, in most practical situations, the adversary cannot obtain the internal information of the victim’s neural network. Furthermore, for reward-based attacks, the agent can perform anomaly detection on the perturbed rewards to detect whether it has been attacked. In this paper, we propose a black-box attack method with corrupted rewards, which employs DRL exploration mechanisms to improve the effectiveness of attacking agents. The adversary builds a deep neural network in advance to learn the successor representation (SR) of each state. Then, the adversary can determine the timing of attacks and generate imperceptible adversarial perturbations based on the values of the SR. Experimental results show that the black-box attack algorithm based on SR proposed in this paper can effectively attack agents with fewer adversarial samples.