arXiv subjects

Yanan Li

Publications and source records attributed to Yanan Li.

At least 55 records · Page 3Linked to original sources

arXiv2023

Q-Net: Query-Informed Few-Shot Medical Image Segmentation

Deep learning has achieved tremendous success in computer vision, while medical image segmentation (MIS) remains a challenge, due to the scarcity of data annotations. Meta-learning techniques for few-shot segmentation (Meta-FSS) have been widely used to tackle this challenge, while they neglect possible distribution shifts between the query image and the support set. In contrast, an experienced clinician can perceive and address such shifts by borrowing information from the query image, then fine-tune or calibrate her prior cognitive model accordingly. Inspired by this, we propose Q-Net, a Query-informed Meta-FSS approach, which mimics in spirit the learning mechanism of an expert clinician. We build Q-Net based on ADNet, a recently proposed anomaly detection-inspired method. Specifically, we add two query-informed computation modules into ADNet, namely a query-informed threshold adaptation module and a query-informed prototype refinement module. Combining them with a dual-path extension of the feature extraction module, Q-Net achieves state-of-the-art performance on widely used abdominal and cardiac magnetic resonance (MR) image datasets. Our work sheds light on a novel way to improve Meta-FSS techniques by leveraging query information.

cs.CV↗

arXiv2023

Hybrid quantum-classical convolutional neural network for phytoplankton classification

The taxonomic composition and abundance of phytoplankton, having direct impact on marine ecosystem dynamic and global environment change, are listed as essential ocean variables. Phytoplankton classification is very crucial for Phytoplankton analysis, but it is very difficult because of the huge amount and tiny volume of Phytoplankton. Machine learning is the principle way of performing phytoplankton image classification automatically. When carrying out large-scale research on the marine phytoplankton, the volume of data increases overwhelmingly and more powerful computational resources are required for the success of machine learning algorithms. Recently, quantum machine learning has emerged as the potential solution for large-scale data processing by harnessing the exponentially computational power of quantum computer. Here, for the first time, we demonstrate the feasibility of quantum deep neural networks for phytoplankton classification. Hybrid quantum-classical convolutional and residual neural networks are developed based on the classical architectures. These models make a proper balance between the limited function of the current quantum devices and the large size of phytoplankton images, which make it possible to perform phytoplankton classification on the near-term quantum computers. Better performance is obtained by the quantum-enhanced models against the classical counterparts. In particular, quantum models converge much faster than classical ones. The present quantum models are versatile, and can be applied for various tasks of image classification in the field of marine science.

quant-ph↗

arXiv2023

Charge Density Wave Order in Superconducting Topological Insulator Nbx-Bi2Se3

Spontaneously broken symmetry in the ground state of Bi-based topological materials can lead to promising Topological superconductors, such as Cu-Bi2Se3, Sr-Bi2Se3 and Nb-Bi2Se3. In recent studies, coexistence of multiple Fermi surface features and superconductivity was found in Nb-Bi2Se3. However, the resultant mutliple Fermi surface--charge density wave has not been experimentally reported yet. In this paper, we report possible evidence of co-occurence of a charge density wave (CDW) ground state and a superconducting (SC) ground state in Nb intercalated Bi2Se3. An intercalation induced Periodic Lattice Distortion appears to help stabilize the CDW state, possibly assisting in the formation of 1D chains and rendering electronic and phonon anisotropy to this intriguing system.

cond-mat.supr-con↗

arXiv2023

Quantum Recurrent Neural Networks for Sequential Learning

Quantum neural network (QNN) is one of the promising directions where the near-term noisy intermediate-scale quantum (NISQ) devices could find advantageous applications against classical resources. Recurrent neural networks are the most fundamental networks for sequential learning, but up to now there is still a lack of canonical model of quantum recurrent neural network (QRNN), which certainly restricts the research in the field of quantum deep learning. In the present work, we propose a new kind of QRNN which would be a good candidate as the canonical QRNN model, where, the quantum recurrent blocks (QRBs) are constructed in the hardware-efficient way, and the QRNN is built by stacking the QRBs in a staggered way that can greatly reduce the algorithm's requirement with regard to the coherent time of quantum devices. That is, our QRNN is much more accessible on NISQ devices. Furthermore, the performance of the present QRNN model is verified concretely using three different kinds of classical sequential data, i.e., meteorological indicators, stock price, and text categorization. The numerical experiments show that our QRNN achieves much better performance in prediction (classification) accuracy against the classical RNN and state-of-the-art QNN models for sequential learning, and can predict the changing details of temporal sequence data. The practical circuit structure and superior performance indicate that the present QRNN is a promising learning model to find quantum advantageous applications in the near term.

quant-ph↗

arXiv2022

Anomalous magneto-thermoelectric behavior in massive Dirac materials

Extensive studies of electron transport in Dirac materials have shown positive magneto-resistance (MR) and positive magneto-thermopower (MTP) in a magnetic field perpendicular to the excitation current or thermal gradient. In contrast, measurements of electron transport often show a negative longitudinal MR and negative MTP for a magnetic field oriented along the excitation current or thermal gradient; this is attributed to the chiral anomaly in Dirac materials. Here, we report a very different magneto-thermoelectric transport behavior in the massive Dirac material ZrTe5. Although thin flakes show a commonly observed positive MR in a perpendicular magnetic field, distinct from other Dirac materials, we observe a sharp negative MTP. In a parallel magnetic field, we still observe a negative longitudinal MR, however, a remarkable positive MTP is observed for the fields parallel to the thermal gradients. Our theoretical calculations suggest that this anomalous magneto-thermoelectric behavior can be attributed to the screened Coulomb scattering. This work demonstrates the significance of impurity scattering in the electron transport of topological materials and provides deep insight into the novel magneto-transport phenomena in Dirac materials.

cond-mat.mtrl-sci↗

arXiv2022

Efficient Single-Image Depth Estimation on Mobile Devices, Mobile AI & AIM 2022 Challenge: Report

Various depth estimation models are now widely used on many mobile and IoT devices for image segmentation, bokeh effect rendering, object tracking and many other mobile tasks. Thus, it is very crucial to have efficient and accurate depth estimation models that can run fast on low-power mobile chipsets. In this Mobile AI challenge, the target was to develop deep learning-based single image depth estimation solutions that can show a real-time performance on IoT platforms and smartphones. For this, the participants used a large-scale RGB-to-depth dataset that was collected with the ZED stereo camera capable to generated depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the Raspberry Pi 4 platform, where the developed solutions were able to generate VGA resolution depth maps at up to 27 FPS while achieving high fidelity results. All models developed in the challenge are also compatible with any Android or Linux-based mobile devices, their detailed description is provided in this paper.

cs.CV↗

arXiv2022

Realistic Bokeh Effect Rendering on Mobile GPUs, Mobile AI & AIM 2022 challenge: Report

As mobile cameras with compact optics are unable to produce a strong bokeh effect, lots of interest is now devoted to deep learning-based solutions for this task. In this Mobile AI challenge, the target was to develop an efficient end-to-end AI-based bokeh effect rendering approach that can run on modern smartphone GPUs using TensorFlow Lite. The participants were provided with a large-scale EBB! bokeh dataset consisting of 5K shallow / wide depth-of-field image pairs captured using the Canon 7D DSLR camera. The runtime of the resulting models was evaluated on the Kirin 9000's Mali GPU that provides excellent acceleration results for the majority of common deep learning ops. A detailed description of all models developed in this challenge is provided in this paper.

eess.IV↗

arXiv2022

Rotational symmetry breaking in superconducting nickelate Nd0.8Sr0.2NiO2 films

The infinite-layer nickelates, isostructural to the high-Tc superconductor cuprates, have risen as a promising platform to host unconventional superconductivity and stimulated growing interests in the condensed matter community. Despite numerous researches, the superconducting pairing symmetry of the nickelate superconductors, the fundamental characteristic of a superconducting state, is still under debate. Moreover, the strong electronic correlation in the nickelates may give rise to a rich phase diagram, where the underlying interplay between the superconductivity and other emerging quantum states with broken symmetry is awaiting exploration. Here, we study the angular dependence of the transport properties on the infinite-layer nickelate Nd0.8Sr0.2NiO2 superconducting films with Corbino-disk configuration. The azimuthal angular dependence of the magnetoresistance (R(ϕ)) manifests the rotational symmetry breaking from isotropy to four-fold (C4) anisotropy with increasing magnetic field, revealing a symmetry breaking phase transition. Approaching the low temperature and large magnetic field regime, an additional two-fold (C2) symmetric component in the R(ϕ) curves and an anomalous upturn of the temperature-dependent critical field are observed simultaneously, suggesting the emergence of an exotic electronic phase. Our work uncovers the evolution of the quantum states with different rotational symmetries and provides deep insight into the global phase diagram of the nickelate superconductors.

cond-mat.supr-con↗

arXiv2022

Alleviating the Sample Selection Bias in Few-shot Learning by Removing Projection to the Centroid

Few-shot learning (FSL) targets at generalization of vision models towards unseen tasks without sufficient annotations. Despite the emergence of a number of few-shot learning methods, the sample selection bias problem, i.e., the sensitivity to the limited amount of support data, has not been well understood. In this paper, we find that this problem usually occurs when the positions of support samples are in the vicinity of task centroid -- the mean of all class centroids in the task. This motivates us to propose an extremely simple feature transformation to alleviate this problem, dubbed Task Centroid Projection Removing (TCPR). TCPR is applied directly to all image features in a given task, aiming at removing the dimension of features along the direction of the task centroid. While the exact task centroid cannot be accurately obtained from limited data, we estimate it using base features that are each similar to one of the support features. Our method effectively prevents features from being too close to the task centroid. Extensive experiments over ten datasets from different domains show that TCPR can reliably improve classification accuracy across various feature extractors, training algorithms and datasets. The code has been made available at https://github.com/KikimorMay/FSL-TCBR.

cs.CV↗

arXiv2022

Towards Efficient and Stable K-Asynchronous Federated Learning with Unbounded Stale Gradients on Non-IID Data

Federated learning (FL) is an emerging privacy-preserving paradigm that enables multiple participants collaboratively to train a global model without uploading raw data. Considering heterogeneous computing and communication capabilities of different participants, asynchronous FL can avoid the stragglers effect in synchronous FL and adapts to scenarios with vast participants. Both staleness and non-IID data in asynchronous FL would reduce the model utility. However, there exists an inherent contradiction between the solutions to the two problems. That is, mitigating the staleness requires to select less but consistent gradients while coping with non-IID data demands more comprehensive gradients. To address the dilemma, this paper proposes a two-stage weighted $K$ asynchronous FL with adaptive learning rate (WKAFL). By selecting consistent gradients and adjusting learning rate adaptively, WKAFL utilizes stale gradients and mitigates the impact of non-IID data, which can achieve multifaceted enhancement in training speed, prediction accuracy and training stability. We also present the convergence analysis for WKAFL under the assumption of unbounded staleness to understand the impact of staleness and non-IID data. Experiments implemented on both benchmark and synthetic FL datasets show that WKAFL has better overall performance compared to existing algorithms.

cs.LG↗

arXiv2022

Continuity of the attractors in time-dependent spaces and applications

In this paper, we investigate the continuity of the attractors in time-dependent phase spaces. (i) We establish two abstract criteria on the upper semicontinuity and the residual continuity of the pullback $\mathscr D$-attractor with respect to the perturbations, and an equivalence criterion between their continuity and the pullback equi-attraction, which generalize the continuity theory of attractors developed recently in [27,28] to that in time-dependent spaces. (ii) We propose the notion of pullback $\mathscr D$-exponential attractor, which includes the notion of time-dependent exponential attractor [33] as its spacial case, and establish its existence and Hölder continuity criterion via quasi-stability method introduced originally by Chueshov and Lasiecka [12,13]. (iii) We apply above-mentioned criteria to the semilinear damped wave equations with perturbed time-dependent speed of propagation: $\eρ(t) u_{tt}+αu_t -Δu+f(u)=g$, with perturbation parameter $\e\in(0, 1]$, to realize above mentioned continuity of pullback $\mathscr D$ and $\mathscr D$-exponential attractors in time-dependent phase spaces, and the method developed here allows to overcome the difficulty of the hyperbolicity of the model. These results deepen and extend recent theory of attractors in time-dependent spaces in literatures [15,20,19].

math.AP↗

arXiv2022

Dual Path Structural Contrastive Embeddings for Learning Novel Objects

Learning novel classes from a very few labeled samples has attracted increasing attention in machine learning areas. Recent research on either meta-learning based or transfer-learning based paradigm demonstrates that gaining information on a good feature space can be an effective solution to achieve favorable performance on few-shot tasks. In this paper, we propose a simple but effective paradigm that decouples the tasks of learning feature representations and classifiers and only learns the feature embedding architecture from base classes via the typical transfer-learning training strategy. To maintain both the generalization ability across base and novel classes and discrimination ability within each class, we propose a dual path feature learning scheme that effectively combines structural similarity with contrastive feature construction. In this way, both inner-class alignment and inter-class uniformity can be well balanced, and result in improved performance. Experiments on three popular benchmarks show that when incorporated with a simple prototype based classifier, our method can still achieve promising results for both standard and generalized few-shot problems in either an inductive or transductive inference setting.

cs.CV↗

arXiv2021

Class-Incremental Few-Shot Object Detection

Conventional detection networks usually need abundant labeled training samples, while humans can learn new concepts incrementally with just a few examples. This paper focuses on a more challenging but realistic class-incremental few-shot object detection problem (iFSD). It aims to incrementally transfer the model for novel objects from only a few annotated samples without catastrophically forgetting the previously learned ones. To tackle this problem, we propose a novel method LEAST, which can transfer with Less forgetting, fEwer training resources, And Stronger Transfer capability. Specifically, we first present the transfer strategy to reduce unnecessary weight adaptation and improve the transfer capability for iFSD. On this basis, we then integrate the knowledge distillation technique using a less resource-consuming approach to alleviate forgetting and propose a novel clustering-based exemplar selection process to preserve more discriminative features previously learned. Being a generic and effective method, LEAST can largely improve the iFSD performance on various benchmarks.

cs.CV↗

arXiv2021

Black-Box Quantum State Preparation with Inverse Coefficients

Black-box quantum state preparation is a fundamental building block for many higher-level quantum algorithms, which is applied to transduce the data from computational basis into amplitude. Here we present a new algorithm for performing black-box state preparation with inverse coefficients based on the technique of inequality test. This algorithm can be used as a subroutine to perform the controlled rotation stage of the Harrow-Hassidim-Lloyd (HHL) algorithm and the associated matrix inversion algorithms with exceedingly low cost. Furthermore, we extend this approach to address the general black-box state preparation problem where the transduced coefficient is a general non-linear function. The present algorithm greatly relieves the need to do arithmetic and the error is only resulted from the truncated error of binary string. It is expected that our algorithm will find wide usage both in the NISQ and fault-tolerant quantum algorithms.

quant-ph↗

arXiv2021

From Cloud to Edge: A First Look at Public Edge Platforms

Public edge platforms have drawn increasing attention from both academia and industry. In this study, we perform a first-of-its-kind measurement study on a leading public edge platform that has been densely deployed in China. Based on this measurement, we quantitatively answer two critical yet unexplored questions. First, from end users' perspective, what is the performance of commodity edge platforms compared to cloud, in terms of the end-to-end network delay, throughput, and the application QoE. Second, from the edge service provider's perspective, how are the edge workloads different from cloud, in terms of their VM subscription, monetary cost, and resource usage. Our study quantitatively reveals the status quo of today's public edge platforms, and provides crucial insights towards developing and operating future edge services.

cs.NI↗

arXiv2021

Lens-free Optical Detection of Thermal Motion of a Sub-millimeter Sphere Diamagnetically Levitated in High Vacuum

Levitated oscillators with millimeter or sub-millimeter size are particularly attractive due to their potential role in studying various fundamental problems and practical applications. One of the crucial issues towards these goals is to achieve efficient measurements of oscillator motion, while this remains a challenge. Here we theoretically propose a lens-free optical detection scheme, which can be used to detect the motion of a millimeter or sub-millimeter levitated oscillator with a measurement efficiency close to the standard quantum limit with a modest optical power. We demonstrate experimentally this scheme on a 0.5 mm diameter micro-sphere that is diamagnetically levitated under high vacuum and room temperature, and the thermal motion is detected with high precision. Based on this system, an estimated acceleration sensitivity of $9.7 \times 10^{-10}\rm g/\sqrt{Hz}$ is achieved, which is more than one order improvement over the best value reported by the levitated mechanical system. Due to the stability of the system, the minimum resolved acceleration of $3.5\times 10^{-12}\rm g$ is reached with measurement times of $10^5$ s. This result is expected to have potential applications in the study of exotic interactions in the millimeter or sub-millimeter range and the realization of compact gravimeter and accelerometer.

quant-ph↗

arXiv2021

Exponential attractor for the viscoelastic wave model with time-dependent memory kernels

The paper is concerned with the exponential attractors for the viscoelastic wave model in $Ω\subset \mathbb R^3$: $$u_{tt}-h_t(0)Δu-\int_0^\infty\partial_sh_t(s)Δu(t-s)\mathrm ds+f(u)=h,$$ with time-dependent memory kernel $h_t(\cdot)$ which is used to model aging phenomena of the material. Conti et al [Amer. J. Math., 2018] recently provided the correct mathematical setting for the model and a well-posedness result within the novel theory of dynamical systems acting on. time-dependent spaces, recently established by Conti, Pata and Temam [J. Differential Equations, 2013], and proved the existence and the regularity of the time-dependent global attractor. In this work, we further study the existence of the time-dependent exponential attractors as well as their regularity. We establish an abstract existence criterion via quasi-stability method introduced originally by Chueshov and Lasiecka [J. Dynam. Diff.Eqs.,2004], and on the basis of the theory and technique developed in [Amer. J. Math., 2018] we further provide a new method to overcome the difficulty of the lack of further regularity to show the existence of the time-dependent exponential attractor. And these techniques can be used to tackle other hyperbolic models.

math.AP↗

arXiv2020

Induced anomalous Hall effect of massive Dirac fermions in ZrTe5 and HfTe5 thin flakes

Researches on anomalous Hall effect (AHE) have been lasting for a century to make clear the underlying physical mechanism. Generally, the AHE appears in magnetic materials, in which extrinsic process related to scattering effects and intrinsic contribution connected with Berry curvature are crucial. Recently, AHE has been counterintuitively observed in non-magnetic topological materials and attributed to the existence of Weyl points. However, the Weyl point scenario would lead to unsaturated AHE even in large magnetic fields and contradicts the saturation of AHE in several tesla (T) in experiments. In this work, we investigate the Hall effect of ZrTe5 and HfTe5 thin flakes in static ultrahigh magnetic fields up to 33 T. We find the AHE saturates to 55 (70) Ohm^-1*cm^-1 for ZrTe5 (HfTe5) thin flakes above ~ 10 T. Combining detailed magnetotransport experiments and Berry curvature calculations, we clarify that the splitting of massive Dirac bands without Weyl points can be responsible for AHE in non-magnetic topological materials ZrTe5 and HfTe5 thin flakes. This model can identify our thin flake samples to be weak topological insulators and serve as a new tool to probe the band structure topology in topological materials.

cond-mat.mtrl-sci↗