Voter model on adaptive networks

Voter model is an important basic model in statistical physics. In recent years, it has been more and more used to describe the process of opinion formation in sociophysics. In real complex systems, the interactive network of individuals is dynamically adjusted, and the evolving network topology and individual behaviors affect each other. Therefore, we propose a linking dynamics to describe the coevolution of network topology and individual behaviors in this paper, and study the voter model on the adaptive network. We theoretically analyze the properties of the voter model, including consensus probability and time. The evolution of opinions on dynamic networks is further analyzed from the perspective of evolutionary game. Finally, a case study of real data is shown to verify the effectiveness of the theory.

Stochastic Simulation Algorithm for effective spreading dynamics on Time-evolving Adaptive NetworX (SSATAN-X)

Modelling and simulating the dynamics of pathogen spreading has been proven crucial to inform public heath decisions, containment strategies, as well as cost-effectiveness calculations. Pathogen spreading is often modelled as a stochastic process that is driven by pathogen exposure on time-evolving contact networks. In adaptive networks, the spreading process depends not only on the dynamics of a contact network, but vice versa, infection dynamics may alter risk behaviour and thus feed back onto contact dynamics, leading to emergent complex dynamics. However, stochastic simulation of pathogen spreading processes on adaptive networks is currently computationally prohibitive. In this manuscript, we propose SSATAN-X, a new algorithm for the accurate stochastic simulation of pathogen spreading on adaptive networks. The key idea of SSATAN-X is to only capture the contact dynamics that are relevant to the spreading process. We show that SSATAN-X captures the contact dynamics and consequently the spreading dynamics accurately. The algorithm achieves a > 10 fold speed-up over the state-of-art stochastic simulation algorithm (SSA). The speed-up with SSATAN-X further increases when the contact dynamics are fast in relation to the spreading process, i.e. if contacts are short-lived and per-exposure infection risks are small, as applicable to most infectious diseases. We envision that SSATAN-X may extend the scope of analysis of pathogen spreading on adaptive networks. Moreover, it may serve to create benchmark data sets to validate novel numerical approaches for simulation, or for the data-driven analysis of the spreading dynamics on adaptive networks. A C++ implementation of the algorithm is available https://github.com/nmalysheva/SSATAN-X

Skeleton-Based Action Recognition Based on Distance Vector and Multihigh View Adaptive Networks

Skeleton-based human action recognition has attracted much attention in the field of computer vision. Most of the previous studies are based on fixed skeleton graphs so that only the local physical dependencies among joints can be captured, resulting in the omission of implicit joint correlations. In addition, under different views, the content of the same action is very different. In some views, keypoints will be blocked, which will cause recognition errors. In this paper, an action recognition method based on distance vector and multihigh view adaptive network (DV-MHNet) is proposed to address this challenging task. Among the mentioned techniques, the multihigh (MH) view adaptive networks are constructed to automatically determine the best observation view at different heights, obtain complete keypoints information of the current frame image, and enhance the robustness and generalization of the model to recognize actions at different heights. Then, the distance vector (DV) mechanism is introduced on this basis to establish the relative distance and relative orientation between different keypoints in the same frame and the same keypoints in different frame to obtain the global potential relationship of each keypoint, and finally by constructing the spatial temporal graph convolutional network to take into account the information in space and time, the characteristics of the action are learned. This paper has done the ablation study with traditional spatial temporal graph convolutional networks and with or without multihigh view adaptive networks, which reasonably proves the effectiveness of the model. The model is evaluated on two widely used action recognition benchmarks (NTU-RGB + D and PKU-MMD). Our method achieves better performance on both datasets.