arXiv subjects

Yuliang Gu

Publications and source records attributed to Yuliang Gu.

21 records · Page 2Linked to original sources

arXiv2023

Multi-population opinion dynamics model

We introduce multi-population opinion dynamics models linked to the bounded confidence model, aiming to explore how interactions between individuals contribute to the emergence of consensus, polarization, or fragmentation. Existing models either neglect agent similarities, sacrificing accuracy for scalability, or prioritize accuracy by introducing agent-wise connections, constraining scalability. Our proposed model captures similarities between agents in scalable matter. In our setting, agents similarities are defined by their group affiliations. Specifically, each sub-population is characterized by its distribution, and the closeness between two sub-populations is measured by the Wasserstein distance of their corresponding distributions. This leads to two mutually connected dynamics: micro, the individual-based dynamics, and the macro, the distribution-based one. The individual-wise interactions take into account the population-wise interactions (similarities), and the population-wise interactions are updated based on the individual-wise interactions. We have proven the well-posedness of our models. Additionally, we conducted several simulations to mimic certain complex social phenomena.

math.DS↗

arXiv2022

Phy-Taylor: Physics-Model-Based Deep Neural Networks

Purely data-driven deep neural networks (DNNs) applied to physical engineering systems can infer relations that violate physics laws, thus leading to unexpected consequences. To address this challenge, we propose a physics-model-based DNN framework, called Phy-Taylor, that accelerates learning compliant representations with physical knowledge. The Phy-Taylor framework makes two key contributions; it introduces a new architectural Physics-compatible neural network (PhN), and features a novel compliance mechanism, we call {\em Physics-guided Neural Network Editing\}. The PhN aims to directly capture nonlinearities inspired by physical quantities, such as kinetic energy, potential energy, electrical power, and aerodynamic drag force. To do so, the PhN augments neural network layers with two key components: (i) monomials of Taylor series expansion of nonlinear functions capturing physical knowledge, and (ii) a suppressor for mitigating the influence of noise. The neural-network editing mechanism further modifies network links and activation functions consistently with physical knowledge. As an extension, we also propose a self-correcting Phy-Taylor framework that introduces two additional capabilities: (i) physics-model-based safety relationship learning, and (ii) automatic output correction when violations of safety occur. Through experiments, we show that (by expressing hard-to-learn nonlinearities directly and by constraining dependencies) Phy-Taylor features considerably fewer parameters, and a remarkably accelerated training process, while offering enhanced model robustness and accuracy.

cs.LG↗

arXiv2022

SL1-Simplex: Safe Velocity Regulation of Self-Driving Vehicles in Dynamic and Unforeseen Environments

This paper proposes a novel extension of the Simplex architecture with model switching and model learning to achieve safe velocity regulation of self-driving vehicles in dynamic and unforeseen environments. To guarantee the reliability of autonomous vehicles, an $\mathcal{L}_{1}$ adaptive controller that compensates for uncertainties and disturbances is employed by the Simplex architecture as a verified safe controller to tolerate concurrent software and physical failures. Meanwhile, safe switching controller is incorporated into the Simplex for safe velocity regulation through the integration of the traction control system and anti-lock braking system. Specifically, the vehicle's angular and longitudinal velocities asymptotically track the provided references that vary with driving environments, while the wheel slips are restricted to safety envelopes to prevent slipping and sliding. Due to the high dependence of vehicle dynamics on the driving environments, the proposed Simplex leverages the finite-time model learning to timely learn and update the vehicle model for $\mathcal{L}_{1}$ adaptive controller, when any deviation from the safety envelope or the uncertainty measurement threshold occurs in the unforeseen driving environments. Finally, the effectiveness of the proposed Simplex architecture for safe velocity regulation is validated by the AutoRally platform.

eess.SY↗