arXiv subjects

Zhiyong Chen

Publications and source records attributed to Zhiyong Chen.

At least 91 records · Page 5Linked to original sources

arXiv2022

Predictive GAN-powered Multi-Objective Optimization for Hybrid Federated Split Learning

As an edge intelligence algorithm for multi-device collaborative training, federated learning (FL) can reduce the communication burden but increase the computing load of wireless devices. In contrast, split learning (SL) can reduce the computing load of devices by using model splitting and assignment, but increase the communication burden to transmit intermediate results. In this paper, to exploit the advantages of FL and SL, we propose a hybrid federated split learning (HFSL) framework in wireless networks, which combines the multi-worker parallel update of FL and flexible splitting of SL. To reduce the computational idleness in model splitting, we design a parallel computing scheme for model splitting without label sharing, and theoretically analyze the influence of the delayed gradient caused by the scheme on the convergence speed. Aiming to obtain the trade-off between the training time and energy consumption, we optimize the splitting decision, the bandwidth and computing resource allocation. The optimization problem is multi-objective, and we thus propose a predictive generative adversarial network (GAN)-powered multi-objective optimization algorithm to obtain the Pareto front of the problem. Experimental results show that the proposed algorithm outperforms others in finding Pareto optimal solutions, and the solutions of the proposed HFSL dominate the solution of FL.

cs.LG↗

arXiv2022

A Sampling Control Framework and Applications to Robust and Adaptive Control

In this paper, we propose a novel sampling control framework based on the emulation technique where the sampling error is regarded as an auxiliary input to the emulated system. Utilizing the supremum norm of sampling error, the design of periodic sampling and event-triggered control law renders the error dynamics bounded-input-bounded-state (BIBS), and when coupled with system dynamics, achieves global or semi-global stabilization. The proposed framework is then extended to tackle the event-triggered and periodic sampling stabilization for a system where only partial state is available for feedback and the system is subject to parameter uncertainties. The proposed framework is further extended to solve two classes of event-triggered adaptive control problems where the emulated closed-loop system does not admit an input-to-state stability (ISS) Lyapunov function. For the first class of systems with linear parameterized uncertainties, even-triggered global adaptive stabilization is achieved without the global Lipschitz condition on nonlinearities as often required in the literature. For the second class of systems with uncertainties whose bound is unknown, the event-triggered adaptive (dynamic) gain controller is designed for the first time. Finally, theoretical results are verified by two numerical examples.

math.OC↗

arXiv2022

Multi-wavelength magnetic coding of helical luminescence in ferromagnetic 2D layered CrI3

Two-dimensional (2D) van der Waals (vdW) ferromagnets have opened new avenues for manipulating spin at the limits of single or few atomic layers, and for creating unique magneto-exciton devices through the coupling of long-range ferromagnetic (FM) orders and excitons. However, 2D vdW ferromagnets explored so far have rarely possessed exciton behaviors; to date, FM CrI3 have been recently revealed to show ligand-field photoluminescence correlated with FM ordering, but typically with a broad emission peak. Alternatively, many-body excitons have been observed in antiferromagnetic (AFM) NiPS3, but the coupling of excitons with AFM orders is exponentially more difficult, owing to extremely high coercivity. Here, we report a straightforward approach to realize strong coupling of narrow helical emission and FM orders at a low magnetic field in CrI3 through a relatively simple microsphere cavity. We show that the resonant whispering-gallery-modes (WGM) of SiO2 microspheres give rising to a series of strong oscillation helical emissions with a full width at half-maximum (FWHM) of ~5 nm under continuous wave excitation. Reversible magnetic control and coding of helical luminescence with multiwavelength is realized in the range of 950-1100 nm. This work enables plenty of opportunities for creating magnetic encoding lasing for photonic integrated chips.

cond-mat.mtrl-sci↗

arXiv2022

Observation of intrinsic crystal phase in bare CrI3 ferromagnetism

Intrinsic structural phase is a crucial foundation for the fundamental physical properties, and for creating innovative devices with unprecedented performances and unique functionalities. Long-range ferromagnetic orders of van der Waals CrI3 are strongly tied with interlayer stacking orders. However, the intrinsic structure of few-layer CrI3 still remains elusive; the predicted monoclinic phase has not yet been experimentally detected in bare few-layer CrI3. Here we uncover the intrinsic structure of few-layer CrI3 with interlayer antiferromagnetic coupling, which unambiguously show monoclinic stacking in both bare and hBN-encapsulated bilayer and tri-five-layer CrI3 throughout an entire temperature range from 300 to 10 K. An exotic spring damping effect from hBN encapsulation layers is experimentally observed in hBN/CrI3/hBN heterostructures, which partly hinders interlayer sliding of CrI3. This work demonstrates the intrinsic monoclinic crystal phase of few-layer CrI3 and associated correlation with magnetic orders, opening up numerous opportunities for creating magnetic texture by stacking design.

cond-mat.mtrl-sci↗

arXiv2022

Fixed-time Synchronization of Networked Uncertain Euler-Lagrange Systems

This paper considers the fixed-time control problem of a multi-agent system composed of a class of Euler-Lagrange dynamics with parametric uncertainty and a dynamic leader under a directed communication network. A distributed fixed-time observer is first proposed to estimate the desired trajectory and then a fixed-time controller is constructed by transforming uncertain Euler-Lagrange systems into second-order systems and utilizing the backstepping design procedure. The overall design guarantees that the synchronization errors converge to zero in a prescribed time independent of initial conditions. The control design conditions can also be relaxed for a weaker finite-time control requirement.

math.OC↗

arXiv2022

Age of Information-based Scheduling for Wireless D2D Systems with a Deep Learning Approach

Device-to-device (D2D) links scheduling for avoiding excessive interference is critical to the success of wireless D2D communications. Most of the traditional scheduling schemes only consider the maximum throughput or fairness of the system and do not consider the freshness of information. In this paper, we propose a novel D2D links scheduling scheme to optimize an age of information (AoI) and throughput jointly scheduling problem when D2D links transmit packets under the last-come-first-serve policy with packet-replacement (LCFS-PR). It is motivated by the fact that the maximum throughput scheduling may reduce the activation probability of links with poor channel conditions, which results in terrible AoI performance. Specifically, We derive the expression of the overall average AoI and throughput of the network under the spatio-temporal interfering queue dynamics with the mean-field assumption. Moreover, a neural network structure is proposed to learn the mapping from the geographic location to the optimal scheduling parameters under a stationary randomized policy, where the scheduling decision can be made without estimating the channel state information(CSI) after the neural network is well-trained. To overcome the problem that implicit loss functions cannot be back-propagated, we derive a numerical solution of the gradient. Finally, numerical results reveal that the performance of the deep learning approach is close to that of a local optimal algorithm which has a higher computational complexity. The trade-off curve of AoI and throughput is also obtained, where the AoI tends to infinity when throughput is maximized.

cs.IT↗

arXiv2022

Adaptive Cooperative Tracking and Parameter Estimation of an Uncertain Leader over General Directed Graphs

This paper studies cooperative tracking problem of heterogeneous Euler-Lagrange systems with an uncertain leader. Different from most existing works, system dynamic knowledge of the leader node is unaccessible to any follower node in our paper. Distributed adaptive observers are designed for all follower nodes, simultaneously estimate the state and parameters of the leader node. The observer design does not rely on the frequency knowledge of the leader node, and the estimation errors are shown to converge to zero exponentially. Moreover, the results are applied to general directed graphs, where the symmetry of Laplacian matrix does not hold. This is due to two newly developed Lyapunov equations, which solely depend on communication network topologies. Interestingly, using these Lyapunov equations, many results of multi-agent systems over undirected graphs can be extended to general directed graphs. Finally, this paper also advances the knowledge base of adaptive control systems by providing a main tool in the analysis of parameter convergence for adaptive observers.

eess.SY↗

arXiv2022

Structural Consensus in Networks with Directed Topologies and Its Cryptographic Implementation

The existing cryptosystem based approaches for privacy-preserving consensus of networked systems are usually limited to those with undirected topologies. This paper proposes a new privacy-preserving algorithm for networked systems with directed topologies to reach confidential consensus. As a prerequisite for applying the algorithm, a structural consensus problem is formulated and the solvability conditions are discussed for an explicitly constructed controller. The controller is then implemented with encryption to achieve consensus while avoiding individual's information leakage to external eavesdroppers and/or malicious internal neighbors.

cs.MA↗

arXiv2021

Coupling effect and pole assignment in trajectory regulation of multi-agent systems

This paper revisits a well studied leader-following consensus problem of linear multi-agent systems, while aiming at follower nodes' transient performance. Conventionally, when not all follower nodes have access to the leader's state information, distributed observers are designed to estimate the leader's state, and the observers are coupled via communication network. Then each follower node only needs to track its observer's state independently, without interacting with its neighbors. This paper deliberately introduces certain coupling effect among follower nodes, such that the follower nodes tend to converge to each other cooperatively on the way they converge to the leader. Moreover, by suitably designing the control law, the poles of follower nodes can be assigned as desired, and thus transient tracking performance can also be adjusted.

math.OC↗

arXiv2021

Optimal Actor-Critic Policy with Optimized Training Datasets

Actor-critic (AC) algorithms are known for their efficacy and high performance in solving reinforcement learning problems, but they also suffer from low sampling efficiency. An AC based policy optimization process is iterative and needs to frequently access the agent-environment system to evaluate and update the policy by rolling out the policy, collecting rewards and states (i.e. samples), and learning from them. It ultimately requires a huge number of samples to learn an optimal policy. To improve sampling efficiency, we propose a strategy to optimize the training dataset that contains significantly less samples collected from the AC process. The dataset optimization is made of a best episode only operation, a policy parameter-fitness model, and a genetic algorithm module. The optimal policy network trained by the optimized training dataset exhibits superior performance compared to many contemporary AC algorithms in controlling autonomous dynamical systems. Evaluation on standard benchmarks show that the method improves sampling efficiency, ensures faster convergence to optima, and is more data-efficient than its counterparts.

cs.LG↗

arXiv2021

Autoencoder-based Semantic Communication Systems with Relay Channels

In this letter, we propose a semantic communication scheme for wireless relay channels based on Autoencoder, named AESC, which encodes and decodes sentences from the semantic dimension. The Autoencoder module provides anti-noise performance for the system. Meanwhile, a novel semantic forward (SF) mode is designed for the relay node to forward the semantic information at the semantic level, especially for the scenarios that there is no common knowledge shared between the source and destination nodes. Numerical results show that the AESC achieves better stability performance than the traditional communication schemes, and the proposed SF mode provides a significant performance gain compared to the traditional forward protocols.

cs.IT↗

arXiv2021

Improved Soft Actor-Critic: Mixing Prioritized Off-Policy Samples with On-Policy Experience

Soft Actor-Critic (SAC) is an off-policy actor-critic reinforcement learning algorithm, essentially based on entropy regularization. SAC trains a policy by maximizing the trade-off between expected return and entropy (randomness in the policy). It has achieved state-of-the-art performance on a range of continuous-control benchmark tasks, outperforming prior on-policy and off-policy methods. SAC works in an off-policy fashion where data are sampled uniformly from past experiences (stored in a buffer) using which parameters of the policy and value function networks are updated. We propose certain crucial modifications for boosting the performance of SAC and make it more sample efficient. In our proposed improved SAC, we firstly introduce a new prioritization scheme for selecting better samples from the experience replay buffer. Secondly we use a mixture of the prioritized off-policy data with the latest on-policy data for training the policy and the value function networks. We compare our approach with the vanilla SAC and some recent variants of SAC and show that our approach outperforms the said algorithmic benchmarks. It is comparatively more stable and sample efficient when tested on a number of continuous control tasks in MuJoCo environments.

cs.LG↗

arXiv2021

Stochastic Optimal Control for Multivariable Dynamical Systems Using Expectation Maximization

Trajectory optimization is a fundamental stochastic optimal control problem. This paper deals with a trajectory optimization approach for dynamical systems subject to measurement noise that can be fitted into linear time-varying stochastic models. Exact/complete solutions to these kind of control problems have been deemed analytically intractable in literature because they come under the category of Partially Observable Markov Decision Processes (POMDPs). Therefore, effective solutions with reasonable approximations are widely sought for. We propose a reformulation of stochastic control in a reinforcement learning setting. This type of formulation assimilates the benefits of conventional optimal control procedure, with the advantages of maximum likelihood approaches. Finally, an iterative trajectory optimization paradigm called as Stochastic Optimal Control - Expectation Maximization (SOC-EM) is put-forth. This trajectory optimization procedure exhibits better performance in terms of reduction of cumulative cost-to-go which is proved both theoretically and empirically. Furthermore, we also provide novel theoretical work which is related to uniqueness of control parameter estimates. Analysis of the control covariance matrix is presented, which handles stochasticity through efficiently balancing exploration and exploitation.

eess.SY↗

arXiv2021

Synchronization of Frequency Modulated Multi-Agent Systems

Oscillation synchronization phenomenon is widely observed in natural systems through frequency modulated signals, especially in biological neural networks. Frequency modulation is also one of most widely used technologies in engineering. However, due to the technical difficulty, oscillations are always simplified as unmodulated sinusoidal-like waves in studying the synchronization mechanism in the bulky literature over the decades. It lacks mathematical principles, especially systems and control theories, for frequency modulated multi-agent systems. This paper aims to bring a new formulation of synchronization of frequency modulated multi-agent systems. It develops new tools to solve the synchronization problem by addressing three issues including frequency observation in nonlinear frequency modulated oscillators subject to network influence, frequency consensus via network interaction subject to observation error, and a well placed small gain condition among them. The architecture of the paper consists of a novel problem formulation, rigorous theoretical development, and numerical verification.

math.OC↗

arXiv2021

IFR: Iterative Fusion Based Recognizer For Low Quality Scene Text Recognition

Although recent works based on deep learning have made progress in improving recognition accuracy on scene text recognition, how to handle low-quality text images in end-to-end deep networks remains a research challenge. In this paper, we propose an Iterative Fusion based Recognizer (IFR) for low quality scene text recognition, taking advantage of refined text images input and robust feature representation. IFR contains two branches which focus on scene text recognition and low quality scene text image recovery respectively. We utilize an iterative collaboration between two branches, which can effectively alleviate the impact of low quality input. A feature fusion module is proposed to strengthen the feature representation of the two branches, where the features from the Recognizer are Fused with image Restoration branch, referred to as RRF. Without changing the recognition network structure, extensive quantitative and qualitative experimental results show that the proposed method significantly outperforms the baseline methods in boosting the recognition accuracy of benchmark datasets and low resolution images in TextZoom dataset.

cs.CV↗

arXiv2021

A Symbolic Regression Method for Dynamic Modeling and Control of Quadrotor UAVs

This paper presents a mathematic dynamic model of a quadrotor unmanned aerial vehicle (QUAV) by using the symbolic regression approach and then proposes a hierarchical control scheme for trajectory tracking. The symbolic regression approach is capable of constructing analytical quadrotor dynamic equations only through the collected data, which relieves the burden of first principle modeling. To facilitate position tracking control of a QUAV, the design of controller can be decomposed into two parts: a proportional-integral controller for the position subsystem is first designed to obtain the desired horizontal position and the backstepping method for the attitude subsystem is developed to ensure that the Euler angles and the altitude can fast converge to the reference values. The effectiveness is verified through experiments on a benchmark multicopter simulator.

eess.SY↗

arXiv2020

Event-triggered controllers based on the supremum norm of sampling-induced error

The paper proposes a novel event-triggered control scheme for nonlinear systems based on the input-delay method. Specifically, the closed-loop system is associated with a pair of auxiliary input and output. The auxiliary output is defined as the derivative of the continuous-time input function, while the auxiliary input is defined as the input disturbance caused by the sampling or equivalently the integral of the auxiliary output over the sampling period. As a result, a cyclic mapping forms from the input to the output via the system dynamics and back from the output to the input via the integral. The event-triggering law is constructed to make the mapping contractive such that the stabilization is achieved and an easy-to-check Zeno-free condition is provided. With this idea, we develop a theorem for the event-triggered control of interconnected nonlinear systems which is employed to solve the event-triggered control for lower-triangular systems with dynamic uncertainties.

eess.SY↗

arXiv2020

Reinforcement Learning Using Expectation Maximization Based Guided Policy Search for Stochastic Dynamics

Guided policy search algorithms have been proven to work with incredible accuracy for not only controlling a complicated dynamical system, but also learning optimal policies from various unseen instances. One assumes true nature of the states in almost all of the well known policy search and learning algorithms. This paper deals with a trajectory optimization procedure for an unknown dynamical system subject to measurement noise using expectation maximization and extends it to learning (optimal) policies which have less noise because of lower variance in the optimal trajectories. Theoretical and empirical evidence of learnt optimal policies of the new approach is depicted in comparison to some well known baselines which are evaluated on an autonomous system with widely used performance metrics.

eess.SY↗