Searcharxiv⌕ Search

arXiv subjects

Qian Zhang

Publications and source records attributed to Qian Zhang.

At least 289 records · Page 16Linked to original sources
arXiv2023

DPSNN: A Differentially Private Spiking Neural Network with Temporal Enhanced Pooling

Privacy protection is a crucial issue in machine learning algorithms, and the current privacy protection is combined with traditional artificial neural networks based on real values. Spiking neural network (SNN), the new generation of artificial neural networks, plays a crucial role in many fields. Therefore, research on the privacy protection of SNN is urgently needed. This paper combines the differential privacy(DP) algorithm with SNN and proposes a differentially private spiking neural network (DPSNN). The SNN uses discrete spike sequences to transmit information, combined with the gradient noise introduced by DP so that SNN maintains strong privacy protection. At the same time, to make SNN maintain high performance while obtaining high privacy protection, we propose the temporal enhanced pooling (TEP) method. It fully integrates the temporal information of SNN into the spatial information transfer, which enables SNN to perform better information transfer. We conduct experiments on static and neuromorphic datasets, and the experimental results show that our algorithm still maintains high performance while providing strong privacy protection.

cs.NE↗
arXiv2023

TinyDet: Accurate Small Object Detection in Lightweight Generic Detectors

Small object detection requires the detection head to scan a large number of positions on image feature maps, which is extremely hard for computation- and energy-efficient lightweight generic detectors. To accurately detect small objects with limited computation, we propose a two-stage lightweight detection framework with extremely low computation complexity, termed as TinyDet. It enables high-resolution feature maps for dense anchoring to better cover small objects, proposes a sparsely-connected convolution for computation reduction, enhances the early stage features in the backbone, and addresses the feature misalignment problem for accurate small object detection. On the COCO benchmark, our TinyDet-M achieves 30.3 AP and 13.5 AP^s with only 991 MFLOPs, which is the first detector that has an AP over 30 with less than 1 GFLOPs; besides, TinyDet-S and TinyDet-L achieve promising performance under different computation limitation.

cs.CV↗
arXiv2023

OpenInst: A Simple Query-Based Method for Open-World Instance Segmentation

Open-world instance segmentation has recently gained significant popularitydue to its importance in many real-world applications, such as autonomous driving, robot perception, and remote sensing. However, previous methods have either produced unsatisfactory results or relied on complex systems and paradigms. We wonder if there is a simple way to obtain state-of-the-art results. Fortunately, we have identified two observations that help us achieve the best of both worlds: 1) query-based methods demonstrate superiority over dense proposal-based methods in open-world instance segmentation, and 2) learning localization cues is sufficient for open world instance segmentation. Based on these observations, we propose a simple query-based method named OpenInst for open world instance segmentation. OpenInst leverages advanced query-based methods like QueryInst and focuses on learning localization cues. Notably, OpenInst is an extremely simple and straightforward framework without any auxiliary modules or post-processing, yet achieves state-of-the-art results on multiple benchmarks. Specifically, in the COCO$\to$UVO scenario, OpenInst achieves a mask AR of 53.3, outperforming the previous best methods by 2.0 AR with a simpler structure. We hope that OpenInst can serve as a solid baselines for future research in this area.

cs.CV↗
arXiv2023

BoxTeacher: Exploring High-Quality Pseudo Labels for Weakly Supervised Instance Segmentation

Labeling objects with pixel-wise segmentation requires a huge amount of human labor compared to bounding boxes. Most existing methods for weakly supervised instance segmentation focus on designing heuristic losses with priors from bounding boxes. While, we find that box-supervised methods can produce some fine segmentation masks and we wonder whether the detectors could learn from these fine masks while ignoring low-quality masks. To answer this question, we present BoxTeacher, an efficient and end-to-end training framework for high-performance weakly supervised instance segmentation, which leverages a sophisticated teacher to generate high-quality masks as pseudo labels. Considering the massive noisy masks hurt the training, we present a mask-aware confidence score to estimate the quality of pseudo masks and propose the noise-aware pixel loss and noise-reduced affinity loss to adaptively optimize the student with pseudo masks. Extensive experiments can demonstrate the effectiveness of the proposed BoxTeacher. Without bells and whistles, BoxTeacher remarkably achieves 35.0 mask AP and 36.5 mask AP with ResNet-50 and ResNet-101 respectively on the challenging COCO dataset, which outperforms the previous state-of-the-art methods by a significant margin and bridges the gap between box-supervised and mask-supervised methods. The code and models will be available at https://github.com/hustvl/BoxTeacher.

cs.CV↗
arXiv2023

Global stability of the Dirac-Klein-Gordon system in two and three space dimensions

In this paper we study global nonlinear stability for the Dirac-Klein-Gordon system in two and three space dimensions for small and regular initial data. In the case of two space dimensions, we consider the Dirac-Klein-Gordon system with a massless Dirac field and a massive scalar field, and prove global existence, sharp time decay estimates and linear scattering for the solutions. In the case of three space dimensions, we consider the Dirac-Klein-Gordon system with a mass parameter in the Dirac equation, and prove uniform (in the mass parameter) global existence, unified time decay estimates and linear scattering in the top order energy space.

math.AP↗
arXiv2023

Global solutions of 2-D cubic Dirac equation with non-compactly supported data

We are interested in the cubic Dirac equation in two space dimensions. We establish the small data global existence and sharp pointwise decay results for general cubic nonlinearities without additional structure. We also prove the scattering of the Dirac equation for certain classes of nonlinearities. In all the above results we do not require the initial data to have compact support.

math.AP↗
arXiv2023

Stability of a coupled wave-Klein-Gordon system with non-compactly supported initial data

In this paper we study global nonlinear stability for a system of semilinear wave and Klein-Gordon equations with quadratic nonlinearities. We consider nonlinearities of the type of wave-Klein-Gordon interactions where there are no derivatives on the wave component. The initial data are assumed to have a suitable polynomial decay at infinity, but are not necessarily compactly supported. We prove small data global existence, sharp pointwise decay estimates and linear scattering results for the solutions. We give a sufficient condition which combines the Sobolev norms of the nonlinearities on flat time slices outside the light cone and the norms of them on truncated hyperboloids inside the light cone, for the linear scattering of the solutions.

math.AP↗
arXiv2023

Nonreciprocal slow or fast light in anti-$\mathcal{PT}$-symmetric optomechanics

Non-Hermitian systems with anti-parity-time ($\mathcal{APT}$) symmetry have revealed rich physics beyond conventional systems. Here, we study optomechanics in an $\mathcal{APT}$-symmetric spinning resonator and show that, by tuning the rotating speed to approach the exceptional point (EP) or the non-Hermitian spectral degeneracy, nonreciprocal light transmission with a high isolation ratio can be realized. Accompanying this process, nonreciprocal group delay or advance is also identified in the vicinity of EP. Our work sheds new light on manipulating laser propagation with optomechanical EP devices and, in a broader view, can be extended to explore a wide range of $\mathcal{APT}$-symmetric effects, such as $\mathcal{APT}$-symmetric phonon lasers, $\mathcal{APT}$-symmetric topological effects, and $\mathcal{APT}$-symmetric force sensing or accelerator.

quant-ph↗
arXiv2023

Complex strengthening mechanisms in nanocrystalline Ni-Mo alloys revealed by a machine-learning interatomic potential

A nanocrystalline metal's strength increases significantly as its grain size decreases, a phenomenon known as the Hall-Petch relation. Such relation, however, breaks down when the grains become too small. Experimental studies have circumvented this problem in a set of Ni-Mo alloys by stabilizing the grain boundaries (GB). Here, using atomistic simulations with a machine learning-based interatomic potential (ML-IAP), we demonstrate that the inverse Hall-Petch relation can be correctly reproduced due to a change in the dominant deformation mechanism as the grain becomes small in the Ni-Mo polycrystals. It is found that the atomic von Mises strain can be significantly reduced by either solute doping and/or annealing for small-grain-size polycrystals, leading to the increased strength of the polycrystals. On the other hand, for large-grain-size polycrystals, annealing weakens the material due to the large atomic movements in GB regions. Over a broad range of grain size, the superposition of the solute and annealing effects on polycrystals enhances the strength of those with small grain size more than those with large ones, giving rise to the continuous strengthening at extremely small grain sizes. Overall, this study not only demonstrates the reliability of the ML-IAP, but also provides a comprehensive atomistic view of complex strengthening mechanisms in nanocrystals, opening up a new avenue to tailor their mechanical properties.

cond-mat.mtrl-sci↗
arXiv2023

Deep-Learning Tool for Early Identifying Non-Traumatic Intracranial Hemorrhage Etiology based on CT Scan

Background: To develop an artificial intelligence system that can accurately identify acute non-traumatic intracranial hemorrhage (ICH) etiology based on non-contrast CT (NCCT) scans and investigate whether clinicians can benefit from it in a diagnostic setting. Materials and Methods: The deep learning model was developed with 1868 eligible NCCT scans with non-traumatic ICH collected between January 2011 and April 2018. We tested the model on two independent datasets (TT200 and SD 98) collected after April 2018. The model's diagnostic performance was compared with clinicians's performance. We further designed a simulated study to compare the clinicians's performance with and without the deep learning system augmentation. Results: The proposed deep learning system achieved area under the receiver operating curve of 0.986 (95% CI 0.967-1.000) on aneurysms, 0.952 (0.917-0.987) on hypertensive hemorrhage, 0.950 (0.860-1.000) on arteriovenous malformation (AVM), 0.749 (0.586-0.912) on Moyamoya disease (MMD), 0.837 (0.704-0.969) on cavernous malformation (CM), and 0.839 (0.722-0.959) on other causes in TT200 dataset. Given a 90% specificity level, the sensitivities of our model were 97.1% and 90.9% for aneurysm and AVM diagnosis, respectively. The model also shows an impressive generalizability in an independent dataset SD98. The clinicians achieve significant improvements in the sensitivity, specificity, and accuracy of diagnoses of certain hemorrhage etiologies with proposed system augmentation. Conclusions: The proposed deep learning algorithms can be an effective tool for early identification of hemorrhage etiologies based on NCCT scans. It may also provide more information for clinicians for triage and further imaging examination selection.

eess.IV↗
arXiv2023

Deep Learning Approach to Predict Hemorrhage in Moyamoya Disease

Objective: Reliable tools to predict moyamoya disease (MMD) patients at risk for hemorrhage could have significant value. The aim of this paper is to develop three machine learning classification algorithms to predict hemorrhage in moyamoya disease. Methods: Clinical data of consecutive MMD patients who were admitted to our hospital between 2009 and 2015 were reviewed. Demographics, clinical, radiographic data were analyzed to develop artificial neural network (ANN), support vector machine (SVM), and random forest models. Results: We extracted 33 parameters, including 11 demographic and 22 radiographic features as input for model development. Of all compared classification results, ANN achieved the highest overall accuracy of 75.7% (95% CI, 68.6%-82.8%), followed by SVM with 69.2% (95% CI, 56.9%-81.5%) and random forest with 70.0% (95% CI, 57.0%-83.0%). Conclusions: The proposed ANN framework can be a potential effective tool to predict the possibility of hemorrhage among adult MMD patients based on clinical information and radiographic features.

cs.LG↗
arXiv2023

Graph Neural Networks Intersect Probabilistic Graphical Models: A Survey

Graphs are a powerful data structure to represent relational data and are widely used to describe complex real-world data structures. Probabilistic Graphical Models (PGMs) have been well-developed in the past years to mathematically model real-world scenarios in compact graphical representations of distributions of variables. Graph Neural Networks (GNNs) are new inference methods developed in recent years and are attracting growing attention due to their effectiveness and flexibility in solving inference and learning problems over graph-structured data. These two powerful approaches have different advantages in capturing relations from observations and how they conduct message passing, and they can benefit each other in various tasks. In this survey, we broadly study the intersection of GNNs and PGMs. Specifically, we first discuss how GNNs can benefit from learning structured representations in PGMs, generate explainable predictions by PGMs, and how PGMs can infer object relationships. Then we discuss how GNNs are implemented in PGMs for more efficient inference and structure learning. In the end, we summarize the benchmark datasets used in recent studies and discuss promising future directions.

cs.AI↗
arXiv2023

MapTR: Structured Modeling and Learning for Online Vectorized HD Map Construction

High-definition (HD) map provides abundant and precise environmental information of the driving scene, serving as a fundamental and indispensable component for planning in autonomous driving system. We present MapTR, a structured end-to-end Transformer for efficient online vectorized HD map construction. We propose a unified permutation-equivalent modeling approach, i.e., modeling map element as a point set with a group of equivalent permutations, which accurately describes the shape of map element and stabilizes the learning process. We design a hierarchical query embedding scheme to flexibly encode structured map information and perform hierarchical bipartite matching for map element learning. MapTR achieves the best performance and efficiency with only camera input among existing vectorized map construction approaches on nuScenes dataset. In particular, MapTR-nano runs at real-time inference speed ($25.1$ FPS) on RTX 3090, $8\times$ faster than the existing state-of-the-art camera-based method while achieving $5.0$ higher mAP. Even compared with the existing state-of-the-art multi-modality method, MapTR-nano achieves $0.7$ higher mAP, and MapTR-tiny achieves $13.5$ higher mAP and $3\times$ faster inference speed. Abundant qualitative results show that MapTR maintains stable and robust map construction quality in complex and various driving scenes. MapTR is of great application value in autonomous driving. Code and more demos are available at \url{https://github.com/hustvl/MapTR}.

cs.CV↗
arXiv2023

Fully H(gradcurl)-nonconforming Finite Element Method for The Singularly Perturbed Quad-curl Problem on Cubical Meshes

In this paper, we develop two fully nonconforming (both H(grad curl)-nonconforming and H(curl)-nonconforming) finite elements on cubical meshes which can fit into the Stokes complex. The newly proposed elements have 24 and 36 degrees of freedom, respectively. Different from the fully H(grad curl)-nonconforming tetrahedral finite elements in [9], the elements in this paper lead to a robust finite element method to solve the singularly perturbed quad-curl problem. To confirm this, we prove the optimal convergence of order $O(h)$ for a fixed parameter $ε$ and the uniform convergence of order $O(h^{1/2})$ for any value of $ε$. Some numerical examples are used to verify the correctness of the theoretical analysis.

math.NA↗
arXiv2022

Towards Accurate Ground Plane Normal Estimation from Ego-Motion

In this paper, we introduce a novel approach for ground plane normal estimation of wheeled vehicles. In practice, the ground plane is dynamically changed due to braking and unstable road surface. As a result, the vehicle pose, especially the pitch angle, is oscillating from subtle to obvious. Thus, estimating ground plane normal is meaningful since it can be encoded to improve the robustness of various autonomous driving tasks (e.g., 3D object detection, road surface reconstruction, and trajectory planning). Our proposed method only uses odometry as input and estimates accurate ground plane normal vectors in real time. Particularly, it fully utilizes the underlying connection between the ego pose odometry (ego-motion) and its nearby ground plane. Built on that, an Invariant Extended Kalman Filter (IEKF) is designed to estimate the normal vector in the sensor's coordinate. Thus, our proposed method is simple yet efficient and supports both camera- and inertial-based odometry algorithms. Its usability and the marked improvement of robustness are validated through multiple experiments on public datasets. For instance, we achieve state-of-the-art accuracy on KITTI dataset with the estimated vector error of 0.39°. Our code is available at github.com/manymuch/ground_normal_filter.

cs.CV↗
arXiv2022

Perceive, Interact, Predict: Learning Dynamic and Static Clues for End-to-End Motion Prediction

Motion prediction is highly relevant to the perception of dynamic objects and static map elements in the scenarios of autonomous driving. In this work, we propose PIP, the first end-to-end Transformer-based framework which jointly and interactively performs online mapping, object detection and motion prediction. PIP leverages map queries, agent queries and mode queries to encode the instance-wise information of map elements, agents and motion intentions, respectively. Based on the unified query representation, a differentiable multi-task interaction scheme is proposed to exploit the correlation between perception and prediction. Even without human-annotated HD map or agent's historical tracking trajectory as guidance information, PIP realizes end-to-end multi-agent motion prediction and achieves better performance than tracking-based and HD-map-based methods. PIP provides comprehensive high-level information of the driving scene (vectorized static map and dynamic objects with motion information), and contributes to the downstream planning and control. Code and models will be released for facilitating further research.

cs.CV↗
arXiv2022

Non-reversible Parallel Tempering for Deep Posterior Approximation

Parallel tempering (PT), also known as replica exchange, is the go-to workhorse for simulations of multi-modal distributions. The key to the success of PT is to adopt efficient swap schemes. The popular deterministic even-odd (DEO) scheme exploits the non-reversibility property and has successfully reduced the communication cost from $O(P^2)$ to $O(P)$ given sufficiently many $P$ chains. However, such an innovation largely disappears in big data due to the limited chains and few bias-corrected swaps. To handle this issue, we generalize the DEO scheme to promote non-reversibility and propose a few solutions to tackle the underlying bias caused by the geometric stopping time. Notably, in big data scenarios, we obtain an appealing communication cost $O(P\log P)$ based on the optimal window size. In addition, we also adopt stochastic gradient descent (SGD) with large and constant learning rates as exploration kernels. Such a user-friendly nature enables us to conduct approximation tasks for complex posteriors without much tuning costs.

cs.LG↗
arXiv2022

SMS: Spiking Marching Scheme for Efficient Long Time Integration of Differential Equations

We propose a Spiking Neural Network (SNN)-based explicit numerical scheme for long time integration of time-dependent Ordinary and Partial Differential Equations (ODEs, PDEs). The core element of the method is a SNN, trained to use spike-encoded information about the solution at previous timesteps to predict spike-encoded information at the next timestep. After the network has been trained, it operates as an explicit numerical scheme that can be used to compute the solution at future timesteps, given a spike-encoded initial condition. A decoder is used to transform the evolved spiking-encoded solution back to function values. We present results from numerical experiments of using the proposed method for ODEs and PDEs of varying complexity.

math.NA↗