arXiv subjects

Guoping Zhao

Publications and source records attributed to Guoping Zhao.

At least 19 recordsLinked to original sources

arXiv2025

Nanofluidic logic based on chiral skyrmion flows

Particle-like chiral magnetic skyrmions can flow in nanotracks and behave like chiral fluids. Using interacting flows to perform logical operations is an important topic in microfluidics and nanofluidics. Here, we report a basic nanofluidic logic computing system based on chiral magnetic skyrmions flowing in parallel pipelines connected by an H-shaped junction. The flow behaviors could be manipulated by adjusting the spin polarization angle, which controls the intrinsic skyrmion Hall angle. We demonstrate that within certain range of the spin polarization angle, fully developed skyrmion flows could lead to fluidic logical operations, which significantly reduce the complexity of skyrmion logic as there is no need for deterministic creation, precise control, and detection of a single isolated skyrmion. Our results suggest that the chiral flow behaviors of magnetic quasiparticles may offer possibilities for spintronic and nanofluidic functions.

cond-mat.mes-hall↗

arXiv2025

Generative Dynamic Graph Representation Learning for Conspiracy Spoofing Detection

Spoofing detection in financial trading is crucial, especially for identifying complex behaviors such as conspiracy spoofing. Traditional machine-learning approaches primarily focus on isolated node features, often overlooking the broader context of interconnected nodes. Graph-based techniques, particularly Graph Neural Networks (GNNs), have advanced the field by leveraging relational information effectively. However, in real-world spoofing detection datasets, trading behaviors exhibit dynamic, irregular patterns. Existing spoofing detection methods, though effective in some scenarios, struggle to capture the complexity of dynamic and diverse, evolving inter-node relationships. To address these challenges, we propose a novel framework called the Generative Dynamic Graph Model (GDGM), which models dynamic trading behaviors and the relationships among nodes to learn representations for conspiracy spoofing detection. Specifically, our approach incorporates the generative dynamic latent space to capture the temporal patterns and evolving market conditions. Raw trading data is first converted into time-stamped sequences. Then we model trading behaviors using the neural ordinary differential equations and gated recurrent units, to generate the representation incorporating temporal dynamics of spoofing patterns. Furthermore, pseudo-label generation and heterogeneous aggregation techniques are employed to gather relevant information and enhance the detection performance for conspiratorial spoofing behaviors. Experiments conducted on spoofing detection datasets demonstrate that our approach outperforms state-of-the-art models in detection accuracy. Additionally, our spoofing detection system has been successfully deployed in one of the largest global trading markets, further validating the practical applicability and performance of the proposed method.

cs.LG↗

arXiv2025

Sharp estimates of the Schr\"{o}dinger type propagators on modulation spaces

This paper is devoted to conducting a comprehensive and self-contained study of the boundedness on modulation spaces of Fourier integral operators arising when solving Schr\"{o}dinger type operators. The symbols of these operators belong to the Sj\"{o}strand class $M^{\infty,1}$, and their phase functions satisfy certain regularity conditions associated with mixed modulation spaces. Our conclusions cover two novel situations corresponding to the so-called mild and high growth of phase functions. These conclusions represent essential improvements and generalizations of existing results. Our method is based on a reasonable decomposition and scaling of the symbol and phase functions, ensuring their membership in appropriate mixed modulation spaces. In a certain sense, all conclusions of this paper are optimal.

math.CA↗

arXiv2025

Extended Hybridization Expansion Solver for Impurity Models with Retarded Interactions

We extend the continuous-time hybridization expansion solver to a general form, where the hybridization function and retarded interaction are treated on equal footing. Correlation functions can now be directly obtained via functional derivatives with respect to the bosonic propagators, similar to the measurement of Green's functions. We devise a combinatorial scheme of measuring the correlation function, whose efficiency partially emulates that of the Green's function measurement. The algorithm and numerical methods are validated through application to an impurity model involving both electron-phonon coupling and exchange interactions, a case where the previous hybridization expansion algorithm is not applicable. Our improvement of the hybridization expansion solver promotes its applicability in studies of electron-phonon coupling, the extended dynamical mean field theory, and the dual boson method.

cond-mat.str-el↗

arXiv2024

Boundedness and compactness kernel theorem for $\alpha$-modulation spaces

This paper is devoted to establishing the kernel theorems for $\alpha$-modulation spaces in terms of boundedness and compactness. We characterize the boundedness of a linear operator $A$ from an $\alpha$-modulation space $M^p_{\alpha}(\mathbb{R}^d)$ into another $\alpha$-modulation space $M^q_{\alpha}(\mathbb{R}^d)$, by the membership of its distributional kernel in mixed $\alpha$-modulation spaces. We also characterize the compactness of $A$ by means of the kernel in a certain closed subspace of mixed $\alpha$-modulation spaces. The proofs are based on the viewpoint that the action of the linear operator on certain function space can be reduced to the action on the suitable atoms of this function space.

math.FA↗

arXiv2024

Transformation of a cellular skyrmion to polyomino-like structures

Topological spin structures with transformable shapes may have potential implications on data storage and computation. Here, we demonstrate that a square cellular skyrmion on an artificial grid pinning pattern can be manipulated by programmed current pulses. We find that parallel short pulses could result in the elongation of the skyrmion mainly in the current direction, while parallel long pulses are able to induce the elongation in the direction perpendicular to the current due to the intrinsic skyrmion Hall effect. Consequently, a programmed sequence of parallel pulses could lead to the transformation of the skyrmion to I-, L-, and Z-shaped polyomino-like structures without affecting the topological charge. In addition, we find that orthogonal pulses could lead to the transformation to more complex polyomino-like structures, including the T-shaped and irregular ones. Particularly, when a small T-shaped structure is formed, the topological charge of the system is found to be non-integer due to incomplete compensation of local topological charge densities; however, the T-shaped structure is stable on the attractive pinning pattern. Our results offer an effective way to create polyomino-like spin structures toward functional applications.

cond-mat.mes-hall↗

arXiv2023

Schr\"{o}dinger operator on Wiener Amalgam spaces in the full range

Using the technique of Gabor analysis, an equivalent characterization is established for the boundedness $e^{i\Delta}: W^{p,q}_m\rightarrow W^{p,q}$, where $0<p_i,q_i\leq\infty$ and $m$ is a $v$-moderate weight. The sharp exponents for the boundedness is also characterized in the case of power weight.

math.CA↗

arXiv2023

Time-Efficient Reinforcement Learning with Stochastic Stateful Policies

Stateful policies play an important role in reinforcement learning, such as handling partially observable environments, enhancing robustness, or imposing an inductive bias directly into the policy structure. The conventional method for training stateful policies is Backpropagation Through Time (BPTT), which comes with significant drawbacks, such as slow training due to sequential gradient propagation and the occurrence of vanishing or exploding gradients. The gradient is often truncated to address these issues, resulting in a biased policy update. We present a novel approach for training stateful policies by decomposing the latter into a stochastic internal state kernel and a stateless policy, jointly optimized by following the stateful policy gradient. We introduce different versions of the stateful policy gradient theorem, enabling us to easily instantiate stateful variants of popular reinforcement learning and imitation learning algorithms. Furthermore, we provide a theoretical analysis of our new gradient estimator and compare it with BPTT. We evaluate our approach on complex continuous control tasks, e.g., humanoid locomotion, and demonstrate that our gradient estimator scales effectively with task complexity while offering a faster and simpler alternative to BPTT.

cs.LG↗

arXiv2023

LocoMuJoCo: A Comprehensive Imitation Learning Benchmark for Locomotion

Imitation Learning (IL) holds great promise for enabling agile locomotion in embodied agents. However, many existing locomotion benchmarks primarily focus on simplified toy tasks, often failing to capture the complexity of real-world scenarios and steering research toward unrealistic domains. To advance research in IL for locomotion, we present a novel benchmark designed to facilitate rigorous evaluation and comparison of IL algorithms. This benchmark encompasses a diverse set of environments, including quadrupeds, bipeds, and musculoskeletal human models, each accompanied by comprehensive datasets, such as real noisy motion capture data, ground truth expert data, and ground truth sub-optimal data, enabling evaluation across a spectrum of difficulty levels. To increase the robustness of learned agents, we provide an easy interface for dynamics randomization and offer a wide range of partially observable tasks to train agents across different embodiments. Finally, we provide handcrafted metrics for each task and ship our benchmark with state-of-the-art baseline algorithms to ease evaluation and enable fast benchmarking.

cs.LG↗

arXiv2023

Chiral Skyrmions Interacting with Chiral Flowers

The chiral nature of active matter plays an important role in the dynamics of active matter interacting with chiral structures. Skyrmions are chiral objects, and their interactions with chiral nanostructures can lead to intriguing phenomena. Here, we explore the random-walk dynamics of a thermally activated chiral skyrmion interacting with a chiral flower-like obstacle in a ferromagnetic layer, which could create topology-dependent outcomes. It is a spontaneous mesoscopic order-from-disorder phenomenon driven by the thermal fluctuations and topological nature of skyrmions that exists only in ferromagnetic and ferrimagnetic systems. The interactions between the skyrmions and chiral flowers at finite temperatures can be utilized to control the skyrmion position and distribution without applying any external driving force or temperature gradient. The phenomenon that thermally activated skyrmions are dynamically coupled to chiral flowers may provide a new way to design topological sorting devices.

cond-mat.mes-hall↗

arXiv2023

Coercivity Mechanisms of Single-Molecule Magnets

Magnetic hysteresis has become a crucial aspect for characterizing single-molecule magnets, but the comprehension of the coercivity mechanism is still a challenge. By using analytical derivation and quantum dynamical simulations, we reveal fundamental rules that govern magnetic relaxation of single molecule magnets under the influence of external magnetic fields, which in turn dictates the hysteresis behavior. Specifically, we find that energy level crossing induced by magnetic fields can drastically increase the relaxation rate and set a coercivity limit. The activation of optical-phonon-mediated quantum tunneling accelerates the relaxation and largely determines the coercivity. Intra-molecular exchange interaction in multi-ion compounds may enhance the coercivity by suppressing key relaxation processes. Unpaired bonding electrons in mixed-valence complexes bear a pre-spin-flip process, which may facilitate magnetization reversal. Underlying these properties are magnetic relaxation processes modulated by the interplay of magnetic fields, phonon spectrum and spin state configuration, which also proposes a fresh perspective for the nearly centurial coercive paradox.

cond-mat.mtrl-sci↗

arXiv2023

Reversible magnetic domain reorientation induced by magnetic field pulses with fixed direction

Nanoscale magnetic domains with controllable configurations could be used for classical and quantum applications, where the switching of magnetization configurations is an essential operation for information processing. Here, we report that the magnetic domain reorientation in a notched ferromagnetic nanotrack can be realized and effectively controlled by applying uniform magnetic field pulses in a fixed in-plane direction perpendicular to the nanotrack. Our micromagnetic simulation results show that the configurations of magnetic domains in the notched nanotrack can be switched between a head-to-head state and a tail-to-tail state in a reversible manner driven by magnetic field pulses, while it is unnecessary to reverse the direction of the magnetic field. Such a unique magnetic domain reorientation dynamics is found to depend on magnetic parameters and nanotrack geometries. The reorientation dynamics of magnetic domains also depends on the strength and length of the applied magnetic field pulse. In addition, we point out that the notches at the center of the nanotrack play an important role for the stabilization of the head-to-head and tail-to-tail states during the magnetic domain reorientation. We also qualitatively explain the field-induced reorientation phenomenon with a simplified two-dimensional macrospin model. Our results may make it possible to build spintronic devices driven by a fixed magnetic field. Our findings may also motivate future studies to investigate the classical and quantum applications based on nanoscale magnetic domains.

cond-mat.mes-hall↗

arXiv2023

Laminar and transiently disordered dynamics of a magnetic skyrmion pipe flow

The world is full of fluids that flow. The fluid nature of flowing skyrmionic quasiparticles is of fundamental physical interest and plays an essential role in the transport of many skyrmions. Here, we report the laminar and transiently disordered dynamic behaviors of many magnetic skyrmions flowing in a pipe channel. The skyrmion flow driven by a uniform current may show a lattice structural transition. The skyrmion flow driven by a non-uniform current shows a dynamically varying lattice structure. A large uniform current could result in the compression of skyrmions toward the channel edge, leading to the transition of the skyrmion pipe flow into an open-channel flow with a free surface. Namely, the width of the skyrmion flow could be adjusted by the driving current. Skyrmions on the free surface may form a single shear layer adjacent to the main skyrmion flow. In addition, although we focus on the skyrmion flow dynamics in a clean pipe channel without any pinning or defect effect, we also show that a variation of magnetic anisotropy in the pipe channel could lead to more complicated skyrmion flow dynamics and pathlines. Our results reveal the fluid nature of skyrmionic quasiparticles that may motivate future research on the complex flow physics of magnetic textures.

cond-mat.mes-hall↗

arXiv2023

LS-IQ: Implicit Reward Regularization for Inverse Reinforcement Learning

Recent methods for imitation learning directly learn a $Q$-function using an implicit reward formulation rather than an explicit reward function. However, these methods generally require implicit reward regularization to improve stability and often mistreat absorbing states. Previous works show that a squared norm regularization on the implicit reward function is effective, but do not provide a theoretical analysis of the resulting properties of the algorithms. In this work, we show that using this regularizer under a mixture distribution of the policy and the expert provides a particularly illuminating perspective: the original objective can be understood as squared Bellman error minimization, and the corresponding optimization problem minimizes a bounded $\chi^2$-Divergence between the expert and the mixture distribution. This perspective allows us to address instabilities and properly treat absorbing states. We show that our method, Least Squares Inverse Q-Learning (LS-IQ), outperforms state-of-the-art algorithms, particularly in environments with absorbing states. Finally, we propose to use an inverse dynamics model to learn from observations only. Using this approach, we retain performance in settings where no expert actions are available.

cs.LG↗

arXiv2022

Reversible Transformation between Isolated Skyrmions and Bimerons

Skyrmions and bimerons are versatile topological spin textures that can be used as information bits for both classical and quantum computing. The transformation between isolated skyrmions and bimerons is an essential operation for computing architecture based on multiple different topological bits. Here we report the creation of isolated skyrmions and their subsequent transformation to bimerons by harnessing the electric current-induced Oersted field and temperature-induced perpendicular magnetic anisotropy variation. The transformation between skyrmions and bimerons is reversible, which is controlled by the current amplitude and scanning direction. Both skyrmions and bimerons can be created in the same system through the skyrmion-bimeron transformation and magnetization switching. Deformed skyrmion bubbles and chiral labyrinth domains are found as nontrivial intermediate transition states. Our results may provide a unique way for building advanced information-processing devices using different types of topological spin textures in the same system.

cond-mat.mes-hall↗

arXiv2022

InvisibiliTee: Angle-agnostic Cloaking from Person-Tracking Systems with a Tee

After a survey for person-tracking system-induced privacy concerns, we propose a black-box adversarial attack method on state-of-the-art human detection models called InvisibiliTee. The method learns printable adversarial patterns for T-shirts that cloak wearers in the physical world in front of person-tracking systems. We design an angle-agnostic learning scheme which utilizes segmentation of the fashion dataset and a geometric warping process so the adversarial patterns generated are effective in fooling person detectors from all camera angles and for unseen black-box detection models. Empirical results in both digital and physical environments show that with the InvisibiliTee on, person-tracking systems' ability to detect the wearer drops significantly.

cs.CV↗

arXiv2022

STAR-GNN: Spatial-Temporal Video Representation for Content-based Retrieval

We propose a video feature representation learning framework called STAR-GNN, which applies a pluggable graph neural network component on a multi-scale lattice feature graph. The essence of STAR-GNN is to exploit both the temporal dynamics and spatial contents as well as visual connections between regions at different scales in the frames. It models a video with a lattice feature graph in which the nodes represent regions of different granularity, with weighted edges that represent the spatial and temporal links. The contextual nodes are aggregated simultaneously by graph neural networks with parameters trained with retrieval triplet loss. In the experiments, we show that STAR-GNN effectively implements a dynamic attention mechanism on video frame sequences, resulting in the emphasis for dynamic and semantically rich content in the video, and is robust to noise and redundancies. Empirical results show that STAR-GNN achieves state-of-the-art performance for Content-Based Video Retrieval.

cs.CV↗

arXiv2022

A note on the operator window of modulation spaces

Inspired by a recent article \cite[JFAA, 28(2):1-34, (2022)]{Skrettingland2022JoFAaA}, this paper is devoted to the study of suitable window class in the framework of bounded linear operators on $L^2(\rd)$. We establish a natural and complete characterization for the window class such that the corresponding STFT leads to equivalent norms of modulation spaces. The positive bounded linear operators are also characterized in Cohen's class distributions such that the corresponding quantities form equivalent norms of modulation spaces. As a generalization, we introduce a family of operator classes corresponding to the operator-valued modulation spaces. Some applications of our main theorems to the localization operators are also concerned.

math.FA↗