arXiv subjects

Xuan Liu

Publications and source records attributed to Xuan Liu.

At least 109 records · Page 6Linked to original sources

arXiv2023

Cautiously-Optimistic Knowledge Sharing for Cooperative Multi-Agent Reinforcement Learning

While decentralized training is attractive in multi-agent reinforcement learning (MARL) for its excellent scalability and robustness, its inherent coordination challenges in collaborative tasks result in numerous interactions for agents to learn good policies. To alleviate this problem, action advising methods make experienced agents share their knowledge about what to do, while less experienced agents strictly follow the received advice. However, this method of sharing and utilizing knowledge may hinder the team's exploration of better states, as agents can be unduly influenced by suboptimal or even adverse advice, especially in the early stages of learning. Inspired by the fact that humans can learn not only from the success but also from the failure of others, this paper proposes a novel knowledge sharing framework called Cautiously-Optimistic kNowledge Sharing (CONS). CONS enables each agent to share both positive and negative knowledge and cautiously assimilate knowledge from others, thereby enhancing the efficiency of early-stage exploration and the agents' robustness to adverse advice. Moreover, considering the continuous improvement of policies, agents value negative knowledge more in the early stages of learning and shift their focus to positive knowledge in the later stages. Our framework can be easily integrated into existing Q-learning based methods without introducing additional training costs. We evaluate CONS in several challenging multi-agent tasks and find it excels in environments where optimal behavioral patterns are difficult to discover, surpassing the baselines in terms of convergence rate and final performance.

cs.MA↗

arXiv2023

XTransCT: Ultra-Fast Volumetric CT Reconstruction using Two Orthogonal X-Ray Projections for Image-guided Radiation Therapy via a Transformer Network

Computed tomography (CT) scans offer a detailed, three-dimensional representation of patients' internal organs. However, conventional CT reconstruction techniques necessitate acquiring hundreds or thousands of x-ray projections through a complete rotational scan of the body, making navigation or positioning during surgery infeasible. In image-guided radiation therapy, a method that reconstructs ultra-sparse X-ray projections into CT images, we can exploit the substantially reduced radiation dose and minimize equipment burden for localization and navigation. In this study, we introduce a novel Transformer architecture, termed XTransCT, devised to facilitate real-time reconstruction of CT images from two-dimensional X-ray images. We assess our approach regarding image quality and structural reliability using a dataset of fifty patients, supplied by a hospital, as well as the larger public dataset LIDC-IDRI, which encompasses thousands of patients. Additionally, we validated our algorithm's generalizability on the LNDb dataset. Our findings indicate that our algorithm surpasses other methods in image quality, structural precision, and generalizability. Moreover, in comparison to previous 3D convolution-based approaches, we note a substantial speed increase of approximately 300 %, achieving 44 ms per 3D image reconstruction.

eess.IV↗

arXiv2023

Integrating Contact-aware Feedback CPG System for Learning-based Soft Snake Robot Locomotion Controllers

This paper aims to solve the contact-aware locomotion problem of a soft snake robot by developing bio-inspired contact-aware locomotion controllers. To provide effective contact information for the controllers, we develop a scale-covered sensor structure mimicking natural snakes' scale sensilla. In the design of the control framework, our core contribution is the development of a novel sensory feedback mechanism for the Matsuoka central pattern generator (CPG) network. This mechanism allows the Matsuoka CPG system to work like a "spinal cord" in the whole contact-aware control scheme, which simultaneously takes the stimuli including tonic input signals from the "brain" (a goal-tracking locomotion controller) and sensory feedback signals from the "reflex arc" (the contact reactive controller), and generates rhythmic signals to actuate the soft snake robot to slither through densely allocated obstacles. In the "reflex arc" design, we develop two distinctive types of reactive controllers -- 1) a reinforcement learning (RL) sensor regulator that learns to manipulate the sensory feedback inputs of the CPG system, and 2) a local reflexive sensor-CPG network that directly connects sensor readings and the CPG's feedback inputs in a specific topology. Combining with the locomotion controller and the Matsuoka CPG system, these two reactive controllers facilitate two different contact-aware locomotion control schemes. The two control schemes are tested and evaluated in both simulated and real soft snake robots, showing promising performance in the contact-aware locomotion tasks. The experimental results also validate the advantages of the modified Matsuoka CPG system with a new sensory feedback mechanism for bio-inspired robot controller design.

cs.RO↗

arXiv2023

Generating HSR Bogie Vibration Signals via Pulse Voltage-Guided Conditional Diffusion Model

Generative Adversarial Networks (GANs) for producing realistic signals, have substantially improved fault diagnosis algorithms in various Internet of Things (IoT) systems. However, challenges such as training instability and dynamical inaccuracy limit their utility in high-speed rail (HSR) bogie fault diagnosis. To address these challenges, we introduce the Pulse Voltage-Guided Conditional Diffusion Model (VGCDM). Unlike traditional implicit GANs, VGCDM adopts a sequential U-Net architecture, facilitating multi-phase denoising diffusion for generation, which bolsters training stability and mitigate convergence issues. VGCDM also incorporates control pulse voltage by cross-attention mechanism to ensure the alignment of vibration with voltage signals, enhancing the Conditional Diffusion Model's progressive controlablity. Consequently, solely straightforward sampling of control voltages, ensuring the efficient transformation from Gaussian Noise to vibration signals. This adaptability remains robust even in scenarios with time-varying speeds. To validate the effectiveness, we conducted two case studies using SQ dataset and high-simulation HSR bogie dataset. The results of our experiments unequivocally confirm that VGCDM outperforms other generative models, achieving the best RSME, PSNR, and FSCS, showing its superiority in conditional HSR bogie vibration signal generation. For access, our code is available at https://github.com/xuanliu2000/VGCDM.

eess.SP↗

arXiv2023

Hierarchical Optimization-Derived Learning

In recent years, by utilizing optimization techniques to formulate the propagation of deep model, a variety of so-called Optimization-Derived Learning (ODL) approaches have been proposed to address diverse learning and vision tasks. Although having achieved relatively satisfying practical performance, there still exist fundamental issues in existing ODL methods. In particular, current ODL methods tend to consider model construction and learning as two separate phases, and thus fail to formulate their underlying coupling and depending relationship. In this work, we first establish a new framework, named Hierarchical ODL (HODL), to simultaneously investigate the intrinsic behaviors of optimization-derived model construction and its corresponding learning process. Then we rigorously prove the joint convergence of these two sub-tasks, from the perspectives of both approximation quality and stationary analysis. To our best knowledge, this is the first theoretical guarantee for these two coupled ODL components: optimization and learning. We further demonstrate the flexibility of our framework by applying HODL to challenging learning tasks, which have not been properly addressed by existing ODL methods. Finally, we conduct extensive experiments on both synthetic data and real applications in vision and other learning tasks to verify the theoretical properties and practical performance of HODL in various application scenarios.

cs.LG↗

arXiv2023

Learning with Constraint Learning: New Perspective, Solution Strategy and Various Applications

The complexity of learning problems, such as Generative Adversarial Network (GAN) and its variants, multi-task and meta-learning, hyper-parameter learning, and a variety of real-world vision applications, demands a deeper understanding of their underlying coupling mechanisms. Existing approaches often address these problems in isolation, lacking a unified perspective that can reveal commonalities and enable effective solutions. Therefore, in this work, we proposed a new framework, named Learning with Constraint Learning (LwCL), that can holistically examine challenges and provide a unified methodology to tackle all the above-mentioned complex learning and vision problems. Specifically, LwCL is designed as a general hierarchical optimization model that captures the essence of these diverse learning and vision problems. Furthermore, we develop a gradient-response based fast solution strategy to overcome optimization challenges of the LwCL framework. Our proposed framework efficiently addresses a wide range of applications in learning and vision, encompassing three categories and nine different problem types. Extensive experiments on synthetic tasks and real-world applications verify the effectiveness of our approach. The LwCL framework offers a comprehensive solution for tackling complex machine learning and computer vision problems, bridging the gap between theory and practice.

cs.CV↗

arXiv2023

Diffusion Probabilistic Priors for Zero-Shot Low-Dose CT Image Denoising

Denoising low-dose computed tomography (CT) images is a critical task in medical image computing. Supervised deep learning-based approaches have made significant advancements in this area in recent years. However, these methods typically require pairs of low-dose and normal-dose CT images for training, which are challenging to obtain in clinical settings. Existing unsupervised deep learning-based methods often require training with a large number of low-dose CT images or rely on specially designed data acquisition processes to obtain training data. To address these limitations, we propose a novel unsupervised method that only utilizes normal-dose CT images during training, enabling zero-shot denoising of low-dose CT images. Our method leverages the diffusion model, a powerful generative model. We begin by training a cascaded unconditional diffusion model capable of generating high-quality normal-dose CT images from low-resolution to high-resolution. The cascaded architecture makes the training of high-resolution diffusion models more feasible. Subsequently, we introduce low-dose CT images into the reverse process of the diffusion model as likelihood, combined with the priors provided by the diffusion model and iteratively solve multiple maximum a posteriori (MAP) problems to achieve denoising. Additionally, we propose methods to adaptively adjust the coefficients that balance the likelihood and prior in MAP estimations, allowing for adaptation to different noise levels in low-dose CT images. We test our method on low-dose CT datasets of different regions with varying dose levels. The results demonstrate that our method outperforms the state-of-the-art unsupervised method and surpasses several supervised deep learning-based methods. Codes are available in https://github.com/DeepXuan/Dn-Dp.

eess.IV↗

arXiv2023

Reinforcement Learning of CPG-regulated Locomotion Controller for a Soft Snake Robot

Intelligent control of soft robots is challenging due to the nonlinear and difficult-to-model dynamics. One promising model-free approach for soft robot control is reinforcement learning (RL). However, model-free RL methods tend to be computationally expensive and data-inefficient and may not yield natural and smooth locomotion patterns for soft robots. In this work, we develop a bio-inspired design of a learning-based goal-tracking controller for a soft snake robot. The controller is composed of two modules: An RL module for learning goal-tracking behaviors given the unmodeled and stochastic dynamics of the robot, and a central pattern generator (CPG) with the Matsuoka oscillators for generating stable and diverse locomotion patterns. We theoretically investigate the maneuverability of Matsuoka CPG's oscillation bias, frequency, and amplitude for steering control, velocity control, and sim-to-real adaptation of the soft snake robot. Based on this analysis, we proposed a composition of RL and CPG modules such that the RL module regulates the tonic inputs to the CPG system given state feedback from the robot, and the output of the CPG module is then transformed into pressure inputs to pneumatic actuators of the soft snake robot. This design allows the RL agent to naturally learn to entrain the desired locomotion patterns determined by the CPG maneuverability. We validated the optimality and robustness of the control design in both simulation and real experiments, and performed extensive comparisons with state-of-art RL methods to demonstrate the benefit of our bio-inspired control design.

cs.RO↗

arXiv2023

A filtered embedded weighted compact nonlinear scheme for hyperbolic conservation law

In situations where a wide range of flow scales are involved, the nonlinear scheme used should be capable of both shock capturing and low-dissipation.Most of the existing WCNS schemes are too dissipative because the weights deviate from ideal weights in the smooth regions caused by small-scale fluctuations. Moreover, due to the defect of the weighting strategy, the two smooth stencils located on the same side of a discontinuity cannot achieve fourth-order when the discontinuity only crosses S0 or S2. In this paper, we proposed the filtered embedded WCNS scheme which is applicable for complex flow simulations involving both shock and small-scale features. In order to overcome the above deficiency of existing WCNS scheme, a new mapping function is proposed to filter the weights deviation out which can map the weights to ideal weights in smooth region. Meanwhile, the embedded process also implemented by this function which is utilized to improve the resolution of shock capturing in certain discontinuity distributions. The approximate-dispersion-relation analysis indicates that the scheme with the mapping function we proposed has lower dispersion error and numerical dissipation as compared to the WCNS-JS and WCNS-Z schemes. The improved performance is demonstrated by the simulation of linear advection problem and nonlinear hyperbolic conservation laws.

physics.comp-ph↗

arXiv2023

Value-Function-based Sequential Minimization for Bi-level Optimization

Gradient-based Bi-Level Optimization (BLO) methods have been widely applied to handle modern learning tasks. However, most existing strategies are theoretically designed based on restrictive assumptions (e.g., convexity of the lower-level sub-problem), and computationally not applicable for high-dimensional tasks. Moreover, there are almost no gradient-based methods able to solve BLO in those challenging scenarios, such as BLO with functional constraints and pessimistic BLO. In this work, by reformulating BLO into approximated single-level problems, we provide a new algorithm, named Bi-level Value-Function-based Sequential Minimization (BVFSM), to address the above issues. Specifically, BVFSM constructs a series of value-function-based approximations, and thus avoids repeated calculations of recurrent gradient and Hessian inverse required by existing approaches, time-consuming especially for high-dimensional tasks. We also extend BVFSM to address BLO with additional functional constraints. More importantly, BVFSM can be used for the challenging pessimistic BLO, which has never been properly solved before. In theory, we prove the asymptotic convergence of BVFSM on these types of BLO, in which the restrictive lower-level convexity assumption is discarded. To our best knowledge, this is the first gradient-based algorithm that can solve different kinds of BLO (e.g., optimistic, pessimistic, and with constraints) with solid convergence guarantees. Extensive experiments verify the theoretical investigations and demonstrate our superiority on various real-world applications.

cs.LG↗

arXiv2023

Global existence and asymptotics for the modified two-dimensional Schrödinger equation in the critical regime

We study the asymptotic behavior of the modified two-dimensional Schrödinger equation $ (D_t -F(D))u=λ|u| u$ in the critical regime, where $λ\in \mathbb{C}$ with $\text{Im} λ\ge0$ and $F(ξ)$ is a second order constant coefficients elliptic symbol. For any smooth initial datum of size $\varepsilon\ll1$, we prove that the solution is global-in-time, combining the vector fields method and a semiclassical analysis method introduced by Delort. Moreover, we present the pointwise decay estimates and the large time asymptotic formulas of the solution.

math.AP↗

arXiv2023

Vibration and jitter of free-flowing thin liquid sheets as target for high-repetition-rate laser-ion acceleration

Very thin free-flowing liquid sheets are promising targets for high-repetition-rate laser-ion acceleration. In this work, we report the generation of micrometer-thin free-flowing liquid sheets from the collision of two liquid jets, and study the vibration and jitter in their surface normal direction. The dependence of their motion amplitudes on the generation parameters is studied in detail. The origins of the vibration and jitter are discussed. Our results indicate that when the generation parameters are optimized, the motion amplitudes in the stable region can be stabilized below 3.7 μm to meet the stringent requirement of sheet position stability for a tight-focusing setup in laser-ion acceleration experiments.

physics.plasm-ph↗

arXiv2022

Surface-specific vibrational spectroscopy of interfacial water reveals large pH change near graphene electrode at low current densities

Molecular-level insight into interfacial water at buried electrode interfaces is essential in elucidating many phenomena of electrochemistry, but spectroscopic probing of the buried interfaces remains challenging. Here, using surface-specific vibrational spectroscopy, we probe and identify the interfacial water orientation and interfacial electric field at the calcium fluoride (CaF2)-supported electrified graphene/water interface under applied potentials. Our data shows that the water orientation changes drastically at negative potentials (<-0.03 V vs. Pd/H2), from O-H group pointing down towards bulk solution to pointing up away from the bulk solution, which arises from charging/discharging not of the graphene but of the CaF2 substrate. The potential-dependent spectra are nearly identical to the pH-dependent spectra, evidencing that the applied potentials change the local pH (more than five pH units) near the graphene electrode even at a current density below 1 microamp per square centimeter. Our work provides molecular-level insights into the dissociation and reorganization of interfacial water on an electrode/electrolyte interface.

physics.chem-ph↗

arXiv2022

Optimization-Derived Learning with Essential Convergence Analysis of Training and Hyper-training

Recently, Optimization-Derived Learning (ODL) has attracted attention from learning and vision areas, which designs learning models from the perspective of optimization. However, previous ODL approaches regard the training and hyper-training procedures as two separated stages, meaning that the hyper-training variables have to be fixed during the training process, and thus it is also impossible to simultaneously obtain the convergence of training and hyper-training variables. In this work, we design a Generalized Krasnoselskii-Mann (GKM) scheme based on fixed-point iterations as our fundamental ODL module, which unifies existing ODL methods as special cases. Under the GKM scheme, a Bilevel Meta Optimization (BMO) algorithmic framework is constructed to solve the optimal training and hyper-training variables together. We rigorously prove the essential joint convergence of the fixed-point iteration for training and the process of optimizing hyper-parameters for hyper-training, both on the approximation quality, and on the stationary analysis. Experiments demonstrate the efficiency of BMO with competitive performance on sparse coding and real-world applications such as image deconvolution and rain streak removal.

cs.LG↗

arXiv2022

Towards Extremely Fast Bilevel Optimization with Self-governed Convergence Guarantees

Gradient methods have become mainstream techniques for Bi-Level Optimization (BLO) in learning and vision fields. The validity of existing works heavily relies on solving a series of approximation subproblems with extraordinarily high accuracy. Unfortunately, to achieve the approximation accuracy requires executing a large quantity of time-consuming iterations and computational burden is naturally caused. This paper is thus devoted to address this critical computational issue. In particular, we propose a single-level formulation to uniformly understand existing explicit and implicit Gradient-based BLOs (GBLOs). This together with our designed counter-example can clearly illustrate the fundamental numerical and theoretical issues of GBLOs and their naive accelerations. By introducing the dual multipliers as a new variable, we then establish Bilevel Alternating Gradient with Dual Correction (BAGDC), a general framework, which significantly accelerates different categories of existing methods by taking specific settings. A striking feature of our convergence result is that, compared to those original unaccelerated GBLO versions, the fast BAGDC admits a unified non-asymptotic convergence theory towards stationarity. A variety of numerical experiments have also been conducted to demonstrate the superiority of the proposed algorithmic framework.

math.OC↗

arXiv2022

Revisiting GANs by Best-Response Constraint: Perspective, Methodology, and Application

In past years, the minimax type single-level optimization formulation and its variations have been widely utilized to address Generative Adversarial Networks (GANs). Unfortunately, it has been proved that these alternating learning strategies cannot exactly reveal the intrinsic relationship between the generator and discriminator, thus easily result in a series of issues, including mode collapse, vanishing gradients and oscillations in the training phase, etc. In this work, by investigating the fundamental mechanism of GANs from the perspective of hierarchical optimization, we propose Best-Response Constraint (BRC), a general learning framework, that can explicitly formulate the potential dependency of the generator on the discriminator. Rather than adopting these existing time-consuming bilevel iterations, we design an implicit gradient scheme with outer-product Hessian approximation as our fast solution strategy. \emph{Noteworthy, we demonstrate that even with different motivations and formulations, a variety of existing GANs ALL can be uniformly improved by our flexible BRC methodology.} Extensive quantitative and qualitative experimental results verify the effectiveness, flexibility and stability of our proposed framework.

cs.LG↗

arXiv2022

Modified scattering for the one-dimensional Schrödinger equation with a subcritical dissipative nonlinearity

We study the asymptotic behavior in time of solutions to the one dimensional nonlinear Schrödinger equation with a subcritical dissipative nonlinearity $λ|u|^αu$, where $0<α<2$, and $λ$ is a complex constant satisfying $\text{Im} λ>\frac{α|\text{Re} λ|}{2\sqrt{ α+1}}$. For arbitrary large initial data, we present the uniform time decay estimates when $4/3<α<2$, and the large time asymptotics of the solution when $\frac{7+\sqrt{145}}{12}<α<2$. The proof is based on the vector fields method and a semiclassical analysis method.

math.AP↗

arXiv2021

Learning to shortcut and shortlist order fulfillment deciding

With the increase of order fulfillment options and business objectives taken into consideration in the deciding process, order fulfillment deciding is becoming more and more complex. For example, with the advent of ship from store retailers now have many more fulfillment nodes to consider, and it is now common to take into account many and varied business goals in making fulfillment decisions. With increasing complexity, efficiency of the deciding process can become a real concern. Finding the optimal fulfillment assignments among all possible ones may be too costly to do for every order especially during peak times. In this work, we explore the possibility of exploiting regularity in the fulfillment decision process to reduce the burden on the deciding system. By using data mining we aim to find patterns in past fulfillment decisions that can be used to efficiently predict most likely assignments for future decisions. Essentially, those assignments that can be predicted with high confidence can be used to shortcut, or bypass, the expensive deciding process, or else a set of most likely assignments can be used for shortlisting -- sending a much smaller set of candidates for consideration by the fulfillment deciding system.

cs.LG↗