arXiv subjects

Rui Wu

Publications and source records attributed to Rui Wu.

At least 73 records · Page 4Linked to original sources

arXiv2023

WIP: Development of a Student-Centered Personalized Learning Framework to Advance Undergraduate Robotics Education

This paper presents a work-in-progress on a learn-ing system that will provide robotics students with a personalized learning environment. This addresses both the scarcity of skilled robotics instructors, particularly in community colleges and the expensive demand for training equipment. The study of robotics at the college level represents a wide range of interests, experiences, and aims. This project works to provide students the flexibility to adapt their learning to their own goals and prior experience. We are developing a system to enable robotics instruction through a web-based interface that is compatible with less expensive hardware. Therefore, the free distribution of teaching materials will empower educators. This project has the potential to increase the number of robotics courses offered at both two- and four-year schools and universities. The course materials are being designed with small units and a hierarchical dependency tree in mind; students will be able to customize their course of study based on the robotics skills they have already mastered. We present an evaluation of a five module mini-course in robotics. Students indicated that they had a positive experience with the online content. They also scored the experience highly on relatedness, mastery, and autonomy perspectives, demonstrating strong motivation potential for this approach.

cs.RO↗

arXiv2023

Multi-scale Transformer Pyramid Networks for Multivariate Time Series Forecasting

Multivariate Time Series (MTS) forecasting involves modeling temporal dependencies within historical records. Transformers have demonstrated remarkable performance in MTS forecasting due to their capability to capture long-term dependencies. However, prior work has been confined to modeling temporal dependencies at either a fixed scale or multiple scales that exponentially increase (most with base 2). This limitation hinders their effectiveness in capturing diverse seasonalities, such as hourly and daily patterns. In this paper, we introduce a dimension invariant embedding technique that captures short-term temporal dependencies and projects MTS data into a higher-dimensional space, while preserving the dimensions of time steps and variables in MTS data. Furthermore, we present a novel Multi-scale Transformer Pyramid Network (MTPNet), specifically designed to effectively capture temporal dependencies at multiple unconstrained scales. The predictions are inferred from multi-scale latent representations obtained from transformers at various scales. Extensive experiments on nine benchmark datasets demonstrate that the proposed MTPNet outperforms recent state-of-the-art methods.

cs.LG↗

arXiv2023

Continuous odor profile monitoring to study olfactory navigation in small animals

Olfactory navigation is observed across species and plays a crucial role in locating resources for survival. In the laboratory, understanding the behavioral strategies and neural circuits underlying odor-taxis requires a detailed understanding of the animal's sensory environment. For small model organisms like C. elegans and larval D. melanogaster, controlling and measuring the odor environment experienced by the animal can be challenging, especially for airborne odors, which are subject to subtle effects from airflow, temperature variation, and from the odor's adhesion, adsorption or reemission. Here we present a method to flexibly control and precisely measure airborne odor concentration in an arena with agar while imaging animal behavior. Crucially and unlike previous methods, our method allows continuous monitoring of the odor profile during behavior. We construct stationary chemical landscapes in an odor flow chamber through spatially patterned odorized air. The odor concentration is measured with a spatially distributed array of digital gas sensors. Careful placement of the sensors allows the odor concentration across the arena to be accurately inferred and continuously monitored at all points in time. We use this approach to measure the precise odor concentration that each animal experiences as it undergoes chemotaxis behavior and report chemotaxis strategies for C. elegans and D. melanogaster larvae populations under different spatial odor landscapes.

q-bio.NC↗

arXiv2022

Probing two driven double quantum dots strongly coupled to a cavity

We experimentally and theoretically study a driven hybrid circuit quantum electrodynamics (cQED) system beyond the dispersive coupling regime. Treating the cavity as part of the driven system, we develop a theory applicable to such strongly coupled and to multi-qubit systems. The fringes measured for a single driven double quantum dot (DQD)-cavity setting and the enlarged splittings of the hybrid Floquet states in the presence of a second DQD are well reproduced with our model. This opens a path to study Floquet states of multi-qubit systems with arbitrarily strong coupling and reveals a new perspective for understanding strongly driven hybrid systems.

cond-mat.mes-hall↗

arXiv2022

Many-body slow quench dynamics and nonadiabatic characterization of topological phases

Previous studies have shown that the bulk topology of single-particle systems can be captured by the band inversion surface or by the spin inversion surface emerged on the time-averaged spin polarization. Most of the studies, however, are based on the single-particle picture even though the systems are fermionic and of multi-bands. Here, we study the many-body quench dynamics of topological systems with all the valence bands fully occupied, and show that the concepts of band inversion surface and spin inversion surface are still valid. More importantly, the many-body quench dynamics is shown to be reduced to a nontrivial three-level Landau-Zener model, which can be solved exactly. Based on the analytical results, the topological spin texture revealed by the time-averaged spin polarization can be applied to characterize the bulk topology and thus provides a direct comparison for future experiments.

cond-mat.str-el↗

arXiv2022

Heterogeneous Information Crossing on Graphs for Session-based Recommender Systems

Recommender systems are fundamental information filtering techniques to recommend content or items that meet users' personalities and potential needs. As a crucial solution to address the difficulty of user identification and unavailability of historical information, session-based recommender systems provide recommendation services that only rely on users' behaviors in the current session. However, most existing studies are not well-designed for modeling heterogeneous user behaviors and capturing the relationships between them in practical scenarios. To fill this gap, in this paper, we propose a novel graph-based method, namely Heterogeneous Information Crossing on Graphs (HICG). HICG utilizes multiple types of user behaviors in the sessions to construct heterogeneous graphs, and captures users' current interests with their long-term preferences by effectively crossing the heterogeneous information on the graphs. In addition, we also propose an enhanced version, named HICG-CL, which incorporates contrastive learning (CL) technique to enhance item representation ability. By utilizing the item co-occurrence relationships across different sessions, HICG-CL improves the recommendation performance of HICG. We conduct extensive experiments on three real-world recommendation datasets, and the results verify that (i) HICG achieves the state-of-the-art performance by utilizing multiple types of behaviors on the heterogeneous graph. (ii) HICG-CL further significantly improves the recommendation performance of HICG by the proposed contrastive learning module.

cs.IR↗

arXiv2022

Conditional Generation Net for Medication Recommendation

Medication recommendation targets to provide a proper set of medicines according to patients' diagnoses, which is a critical task in clinics. Currently, the recommendation is manually conducted by doctors. However, for complicated cases, like patients with multiple diseases at the same time, it's difficult to propose a considerate recommendation even for experienced doctors. This urges the emergence of automatic medication recommendation which can help treat the diagnosed diseases without causing harmful drug-drug interactions.Due to the clinical value, medication recommendation has attracted growing research interests.Existing works mainly formulate medication recommendation as a multi-label classification task to predict the set of medicines. In this paper, we propose the Conditional Generation Net (COGNet) which introduces a novel copy-or-predict mechanism to generate the set of medicines. Given a patient, the proposed model first retrieves his or her historical diagnoses and medication recommendations and mines their relationship with current diagnoses. Then in predicting each medicine, the proposed model decides whether to copy a medicine from previous recommendations or to predict a new one. This process is quite similar to the decision process of human doctors. We validate the proposed model on the public MIMIC data set, and the experimental results show that the proposed model can outperform state-of-the-art approaches.

cs.LG↗

arXiv2022

Commentary: analyzing binary data using MCPMod when zero counts are expected

Bretz et al (2005) proposed multiple Comparison Procedure and Modeling (MCPMod) method to design and analyze dose-finding study. Pinheiro (2014) then generalized it to various types of endpoint, including but not limited to binary endpoint, survival endpoint, count data, and longitudinal data. Pinheiro (2013) recommended to use the estimated covariance matrix from the observed data to recalculate the optimal contrast and the critical value of the test For many phase II studies it is common to have small sample sizes per arm with low placebo response rates jointly. Under such circumstances, it cannot be excluded to have a zero count observed. For example, when the placebo response rate is 10%, there is about 4% chance to observe zero responders in the placebo group, or other dose group(s), which has a similar response rate as placebo. In this manuscript, we would like to illustrate the potential problem of Pinheiro (2013) using a case study and simulations. An alternative method using Firth's logistic regression was evaluated to get a stable estimate of response for each dose group. In addition, we evaluated two options to address the issue with problematic contrast coefficients.

stat.ME↗

arXiv2021

Electrical detection of the spin-flop and room-temperature magnetic ordering in van der Waals CrPS$_{4}$/(Pt, Pd) heterostructures

We study magneto-transport in heterostructures composed of the van der Waals antiferromagnet CrPS$_{4}$ and the heavy metals Pt and Pd. The transverse resistance (R$_{xy}$) signal reveals the spin-flop transition of CrPS$_{4}$ and a strongly enhanced magnetic ordering temperature (>300 K), which might originate from a strong spin-orbit coupling at the interface. While CrPS$_{4}$/Pt devices allow for easy detection of the spin-flop transition, CrPS$_{4}$/Pd devices show a more substantial enhancement in magnetic ordering temperature and exhibit a topological Hall effect signal, possibly related to chiral spin structures at the interface. The longitudinal magnetoresistance (R$_{xx}$) results from a combination of spin-Hall magnetoresistance and the negative magnetoresistance that can be explained by a field-induced change of the electronic band structure of CrPS$_{4}$.

physics.app-ph↗

arXiv2021

HybridGazeNet: Geometric model guided Convolutional Neural Networks for gaze estimation

As a critical cue for understanding human intention, human gaze provides a key signal for Human-Computer Interaction(HCI) applications. Appearance-based gaze estimation, which directly regresses the gaze vector from eye images, has made great progress recently based on Convolutional Neural Networks(ConvNets) architecture and open-source large-scale gaze datasets. However, encoding model-based knowledge into CNN model to further improve the gaze estimation performance remains a topic that needs to be explored. In this paper, we propose HybridGazeNet(HGN), a unified framework that encodes the geometric eyeball model into the appearance-based CNN architecture explicitly. Composed of a multi-branch network and an uncertainty module, HybridGazeNet is trained using a hyridized strategy. Experiments on multiple challenging gaze datasets shows that HybridGazeNet has better accuracy and generalization ability compared with existing SOTA methods. The code will be released later.

cs.CV↗

arXiv2021

Multi-scale 2D Representation Learning for weakly-supervised moment retrieval

Video moment retrieval aims to search the moment most relevant to a given language query. However, most existing methods in this community often require temporal boundary annotations which are expensive and time-consuming to label. Hence weakly supervised methods have been put forward recently by only using coarse video-level label. Despite effectiveness, these methods usually process moment candidates independently, while ignoring a critical issue that the natural temporal dependencies between candidates in different temporal scales. To cope with this issue, we propose a Multi-scale 2D Representation Learning method for weakly supervised video moment retrieval. Specifically, we first construct a two-dimensional map for each temporal scale to capture the temporal dependencies between candidates. Two dimensions in this map indicate the start and end time points of these candidates. Then, we select top-K candidates from each scale-varied map with a learnable convolutional neural network. With a newly designed Moments Evaluation Module, we obtain the alignment scores of the selected candidates. At last, the similarity between captions and language query is served as supervision for further training the candidates' selector. Experiments on two benchmark datasets Charades-STA and ActivityNet Captions demonstrate that our approach achieves superior performance to state-of-the-art results.

cs.CV↗

arXiv2021

You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection

Can Transformer perform 2D object- and region-level recognition from a pure sequence-to-sequence perspective with minimal knowledge about the 2D spatial structure? To answer this question, we present You Only Look at One Sequence (YOLOS), a series of object detection models based on the vanilla Vision Transformer with the fewest possible modifications, region priors, as well as inductive biases of the target task. We find that YOLOS pre-trained on the mid-sized ImageNet-1k dataset only can already achieve quite competitive performance on the challenging COCO object detection benchmark, e.g., YOLOS-Base directly adopted from BERT-Base architecture can obtain 42.0 box AP on COCO val. We also discuss the impacts as well as limitations of current pre-train schemes and model scaling strategies for Transformer in vision through YOLOS. Code and pre-trained models are available at https://github.com/hustvl/YOLOS.

cs.CV↗

arXiv2021

What Makes for Hierarchical Vision Transformer?

Recent studies indicate that hierarchical Vision Transformer with a macro architecture of interleaved non-overlapped window-based self-attention \& shifted-window operation is able to achieve state-of-the-art performance in various visual recognition tasks, and challenges the ubiquitous convolutional neural networks (CNNs) using densely slid kernels. Most follow-up works attempt to replace the shifted-window operation with other kinds of cross-window communication paradigms, while treating self-attention as the de-facto standard for window-based information aggregation. In this manuscript, we question whether self-attention is the only choice for hierarchical Vision Transformer to attain strong performance, and the effects of different kinds of cross-window communication. To this end, we replace self-attention layers with embarrassingly simple linear mapping layers, and the resulting proof-of-concept architecture termed as LinMapper can achieve very strong performance in ImageNet-1k image recognition. Moreover, we find that LinMapper is able to better leverage the pre-trained representations from image recognition and demonstrates excellent transfer learning properties on downstream dense prediction tasks such as object detection and instance segmentation. We also experiment with other alternatives to self-attention for content aggregation inside each non-overlapped window under different cross-window communication approaches, which all give similar competitive results. Our study reveals that the \textbf{macro architecture} of Swin model families, other than specific aggregation layers or specific means of cross-window communication, may be more responsible for its strong performance and is the real challenger to the ubiquitous CNN's dense sliding window paradigm. Code and models will be publicly available to facilitate future research.

cs.CV↗

arXiv2021

Moire Superlattice Modulations in Single-Unit-Cell FeTe Films Grown on NbSe2 Single Crystals

Interface can be a fertile ground for exotic quantum states, including topological superconductivity, Majorana mode, fractal quantum Hall effect, unconventional superconductivity, Mott insulator, etc. Here we grow single-unit-cell (1UC) FeTe film on NbSe2 single crystal by molecular beam epitaxy (MBE) and investigate the film in-situ with home-made cryogenic scanning tunneling microscopy (STM) and non-contact atomic force microscopy (AFM) combined system. We find different stripe-like superlattice modulations on grown FeTe film with different misorientation angles with respect to NbSe2 substrate. We show that these stripe-like superlattice modulations can be understood as moire pattern forming between FeTe film and NbSe2 substrate. Our results indicate that the interface between FeTe and NbSe2 is atomically sharp. By STM-AFM combined measurement, we suggest the moire superlattice modulations have an electronic origin when the misorientation angle is relatively small (<= 3 degree) and have structural relaxation when the misorientation angle is relatively large (>= 10 degree).

cond-mat.str-el↗

arXiv2021

Thermal dynamics of charge density wave pinning in ZrTe3

Impurity pinning has long been discussed to have a profound effect on the dynamics of an incommensurate charge density wave (CDW), which would otherwise slide through the lattice without resistance. Here we visualize the impurity pinning evolution of the CDW in ZrTe3 using the variable temperature scanning tunneling microscopy (STM). At low temperatures, we observe a quasi-1D incommensurate CDW modulation moderately correlated to the impurity positions, indicating a weak impurity pinning. As we raise the sample temperature, the CDW modulation gets progressively weakened and distorted, while the correlation with the impurities becomes stronger. Above the CDW transition temperature, short-range modulations persist with the phase almost all pinned by impurities. The evolution from weak to strong impurity pinning through the CDW transition can be understood as a result of losing phase rigidity.

cond-mat.str-el↗

arXiv2021

Magnetic coupling in Y$_3$Fe$_5$O$_{12}$/Gd$_3$Fe$_5$O$_{12}$ heterostructures

Ferrimagnetic Y$_3$Fe$_5$O$_{12}$ (YIG) is the prototypical material for studying magnonic properties due to its exceptionally low damping. By substituting the yttrium with other rare earth elements that have a net magnetic moment, we can introduce an additional spin degree of freedom. Here, we study the magnetic coupling in epitaxial Y$_3$Fe$_5$O$_{12}$/Gd$_3$Fe$_5$O$_{12}$ (YIG/GIG) heterostructures grown by pulsed laser deposition. From bulk sensitive magnetometry and surface sensitive spin Seebeck effect (SSE) and spin Hall magnetoresistance (SMR) measurements, we determine the alignment of the heterostructure magnetization through temperature and external magnetic field. The ferromagnetic coupling between the Fe sublattices of YIG and GIG dominates the overall behavior of the heterostructures. Due to the temperature dependent gadolinium moment, a magnetic compensation point of the total bilayer system can be identified. This compensation point shifts to lower temperatures with increasing thickness of YIG due the parallel alignment of the iron moments. We show that we can control the magnetic properties of the heterostructures by tuning the thickness of the individual layers, opening up a large playground for magnonic devices based on coupled magnetic insulators. These devices could potentially control the magnon transport analogously to electron transport in giant magnetoresistive devices.

cond-mat.mtrl-sci↗

arXiv2020

Identifying the origin of the non-monotonic thickness dependence of spin-orbit torques and interfacial Dzyaloshinskii-Moriya interaction in a ferrimagnetic insulator heterostructure

Electrical manipulation of magnetism via spin-orbit torques (SOTs) promises efficient spintronic devices. In systems comprising magnetic insulators and heavy metals, SOTs have started to be investigated only recently, especially in systems with interfacial Dzyaloshinskii-Moriya interaction (iDMI). Here, we quantitatively study the SOT efficiency and iDMI in a series of gadolinium gallium garnet (GGG) / thulium iron garnet (TmIG) / platinum (Pt) heterostructures with varying TmIG and Pt thicknesses. We find that the non-monotonic SOT efficiency as a function of the magnetic layer thickness is not consistent with the 1/thickness dependence expected from a simple interfacial SOT mechanism. Moreover, considering the insulating nature of TmIG, our results cannot be explained by the SOT mechanism established for metallic magnets where the transverse charge spin current can inject and dephase in the magnetic layers. Rather we can explain this non-monotonic behavior by a model based on the interfacial spin mixing conductance that is affected by the thickness-dependent exchange splitting energy by determining the phase difference of the reflected spin-up and spin-down electrons at the TmIG / Pt interface. By studying the Pt thickness dependence, we find that the effective DMI for GGG / TmIG / Pt does not depend on the Pt thickness, which indicates that the GGG / TmIG interface is the source of the iDMI in this system. Our work demonstrates that SOT and DMI can originate from two different interfaces, which enables independent optimization of DMI and SOT for advanced chiral spintronics with low damping magnetic insulators.

cond-mat.mtrl-sci↗

arXiv2020

Learning Heatmap-Style Jigsaw Puzzles Provides Good Pretraining for 2D Human Pose Estimation

The target of 2D human pose estimation is to locate the keypoints of body parts from input 2D images. State-of-the-art methods for pose estimation usually construct pixel-wise heatmaps from keypoints as labels for learning convolution neural networks, which are usually initialized randomly or using classification models on ImageNet as their backbones. We note that 2D pose estimation task is highly dependent on the contextual relationship between image patches, thus we introduce a self-supervised method for pretraining 2D pose estimation networks. Specifically, we propose Heatmap-Style Jigsaw Puzzles (HSJP) problem as our pretext-task, whose target is to learn the location of each patch from an image composed of shuffled patches. During our pretraining process, we only use images of person instances in MS-COCO, rather than introducing extra and much larger ImageNet dataset. A heatmap-style label for patch location is designed and our learning process is in a non-contrastive way. The weights learned by HSJP pretext task are utilised as backbones of 2D human pose estimator, which are then finetuned on MS-COCO human keypoints dataset. With two popular and strong 2D human pose estimators, HRNet and SimpleBaseline, we evaluate mAP score on both MS-COCO validation and test-dev datasets. Our experiments show that downstream pose estimators with our self-supervised pretraining obtain much better performance than those trained from scratch, and are comparable to those using ImageNet classification models as their initial backbones.

cs.CV↗