arXiv subjects

Dong Li

Publications and source records attributed to Dong Li.

At least 379 records · Page 21Linked to original sources

arXiv2022

HPS-Det: Dynamic Sample Assignment with Hyper-Parameter Search for Object Detection

Sample assignment plays a prominent part in modern object detection approaches. However, most existing methods rely on manual design to assign positive / negative samples, which do not explicitly establish the relationships between sample assignment and object detection performance. In this work, we propose a novel dynamic sample assignment scheme based on hyper-parameter search. We first define the number of positive samples assigned to each ground truth as the hyper-parameters and employ a surrogate optimization algorithm to derive the optimal choices. Then, we design a dynamic sample assignment procedure to dynamically select the optimal number of positives at each training iteration. Experiments demonstrate that the resulting HPS-Det brings improved performance over different object detection baselines. Moreover, We analyze the hyper-parameter reusability when transferring between different datasets and between different backbones for object detection, which exhibits the superiority and versatility of our method.

cs.CV↗

arXiv2022

Auto Machine Learning for Medical Image Analysis by Unifying the Search on Data Augmentation and Neural Architecture

Automated data augmentation, which aims at engineering augmentation policy automatically, recently draw a growing research interest. Many previous auto-augmentation methods utilized a Density Matching strategy by evaluating policies in terms of the test-time augmentation performance. In this paper, we theoretically and empirically demonstrated the inconsistency between the train and validation set of small-scale medical image datasets, referred to as in-domain sampling bias. Next, we demonstrated that the in-domain sampling bias might cause the inefficiency of Density Matching. To address the problem, an improved augmentation search strategy, named Augmented Density Matching, was proposed by randomly sampling policies from a prior distribution for training. Moreover, an efficient automatical machine learning(AutoML) algorithm was proposed by unifying the search on data augmentation and neural architecture. Experimental results indicated that the proposed methods outperformed state-of-the-art approaches on MedMNIST, a pioneering benchmark designed for AutoML in medical image analysis.

cs.CV↗

arXiv2022

Rethinking Reinforcement Learning based Logic Synthesis

Recently, reinforcement learning has been used to address logic synthesis by formulating the operator sequence optimization problem as a Markov decision process. However, through extensive experiments, we find out that the learned policy makes decisions independent from the circuit features (i.e., states) and yields an operator sequence that is permutation invariant to some extent in terms of operators. Based on these findings, we develop a new RL-based method that can automatically recognize critical operators and generate common operator sequences generalizable to unseen circuits. Our algorithm is verified on both the EPFL benchmark, a private dataset and a circuit at industrial scale. Experimental results demonstrate that it achieves a good balance among delay, area and runtime, and is practical for industrial usage.

cs.LG↗

arXiv2022

CHES: a space-borne astrometric mission for the detection of habitable planets of the nearby solar-type stars

The Closeby Habitable Exoplanet Survey (CHES) mission is proposed to discover habitable-zone Earth-like planets of the nearby solar-type stars ($\sim 10~\mathrm{pc}$ away from our solar system) via micro-arcsecond relative astrometry. The major scientific objectives of CHES are: to search for Earth Twins or terrestrial planets in habitable zones orbiting 100 FGK nearby stars; further to conduct a comprehensive survey and extensively characterize the nearby planetary systems. The primary payload is a high-quality, low-distortion, high-stability telescope. The optical subsystem is a coaxial three-mirror anastigmat (TMA) with a $1.2 \mathrm{~m}$-aperture, $0.44^{\circ} \times 0.44^{\circ}$ field of view and $500 \mathrm{~nm}-900 \mathrm{~nm}$ working waveband. The camera focal plane is composed of 81 MOSAIC scientific CMOS detectors each with $4 \mathrm{~K} \times 4 \mathrm{~K}$ pixels. The heterodyne laser interferometric calibration technology is employed to ensure micro-arcsecond level (1 $μ$as) relative astrometry precision to meet the requirements for detection of Earth-like planets. CHES satellite operates at the Sun-Earth L2 point and observes the entire target stars for 5 years. CHES will offer the first direct measurements of true masses and inclinations of Earth Twins and super-Earths orbiting our neighbor stars based on micro-arcsecond astrometry from space. This will definitely enhance our understanding of the formation of diverse nearby planetary systems and the emergence of other worlds for solar-type stars, and finally to reflect the evolution of our own solar system.

astro-ph.EP↗

arXiv2022

SEREN: Knowing When to Explore and When to Exploit

Efficient reinforcement learning (RL) involves a trade-off between "exploitative" actions that maximise expected reward and "explorative'" ones that sample unvisited states. To encourage exploration, recent approaches proposed adding stochasticity to actions, separating exploration and exploitation phases, or equating reduction in uncertainty with reward. However, these techniques do not necessarily offer entirely systematic approaches making this trade-off. Here we introduce SElective Reinforcement Exploration Network (SEREN) that poses the exploration-exploitation trade-off as a game between an RL agent -- \exploiter, which purely exploits known rewards, and another RL agent -- \switcher, which chooses at which states to activate a pure exploration policy that is trained to minimise system uncertainty and override Exploiter. Using a form of policies known as impulse control, \switcher is able to determine the best set of states to switch to the exploration policy while Exploiter is free to execute its actions everywhere else. We prove that SEREN converges quickly and induces a natural schedule towards pure exploitation. Through extensive empirical studies in both discrete (MiniGrid) and continuous (MuJoCo) control benchmarks, we show that SEREN can be readily combined with existing RL algorithms to yield significant improvement in performance relative to state-of-the-art algorithms.

cs.LG↗

arXiv2022

Quasi-periodic Accelerations of Energetic Particles during a Solar Flare

We report the observation of non-stationary Quasi-Periodic Pulsations (QPPs) in high-energy particles during the impulsive phase of an X4.8 flare on 2002 July 23 (SOL2002-07-23T00:35). The X4.8 flare was simultaneously measured by the Reuven Ramaty High Energy Solar Spectroscopic Imager, Nobeyama Radio Polarimeters, and Nobeyama Radioheliograph. The quasi-period of about 50 s, determined by the wavelet transform, is detected in the Gamma-ray line emission. Using the same method, a quasi-period of about 90 s is found in Gamma-ray continuum, hard X-ray (HXR) and radio emissions during almost the same time. Our observations suggest that the flare QPPs should be associated with energetic ions and nonthermal electrons that quasi-periodically accelerated by the repetitive magnetic reconnection. The different quasi-periods between Gamma-ray line and continuum/HXR/radio emissions indicate an apparent difference in acceleration or propagation between energetic ions and nonthermal electrons of this solar flare.

astro-ph.SR↗

arXiv2022

A disorder-sensitive emergent vortex phase identified in high-Tc superconductor (Li,Fe)OHFeSe

The magneto-transport properties are systematically measured under c-direction fields up to 33 T for a series of single-crystal films of intercalated iron-selenide superconductor (Li,Fe)OHFeSe. The film samples with varying degree of disorder are grown hydrothermally. We observe a magnetic-field-enhanced shoulder-like feature in the mixed state of the high-Tc (Li,Fe)OHFeSe films with weak disorder, while the feature fades away in the films with enhanced disorder. The irreversibility field is significantly suppressed to lower temperatures with the appearance of the shoulder feature. Based on the experiment and model analysis, we establish a new vortex phase diagram for the weakly disordered high-Tc (Li,Fe)OHFeSe, which features an emergent dissipative vortex phase intermediate between the common vortex glass and liquid phases. The reason for the emergence of this intermediate vortex state is further discussed based on related experiments and models.

cond-mat.supr-con↗

arXiv2022

Dynamic Sparse R-CNN

Sparse R-CNN is a recent strong object detection baseline by set prediction on sparse, learnable proposal boxes and proposal features. In this work, we propose to improve Sparse R-CNN with two dynamic designs. First, Sparse R-CNN adopts a one-to-one label assignment scheme, where the Hungarian algorithm is applied to match only one positive sample for each ground truth. Such one-to-one assignment may not be optimal for the matching between the learned proposal boxes and ground truths. To address this problem, we propose dynamic label assignment (DLA) based on the optimal transport algorithm to assign increasing positive samples in the iterative training stages of Sparse R-CNN. We constrain the matching to be gradually looser in the sequential stages as the later stage produces the refined proposals with improved precision. Second, the learned proposal boxes and features remain fixed for different images in the inference process of Sparse R-CNN. Motivated by dynamic convolution, we propose dynamic proposal generation (DPG) to assemble multiple proposal experts dynamically for providing better initial proposal boxes and features for the consecutive training stages. DPG thereby can derive sample-dependent proposal boxes and features for inference. Experiments demonstrate that our method, named Dynamic Sparse R-CNN, can boost the strong Sparse R-CNN baseline with different backbones for object detection. Particularly, Dynamic Sparse R-CNN reaches the state-of-the-art 47.2% AP on the COCO 2017 validation set, surpassing Sparse R-CNN by 2.2% AP with the same ResNet-50 backbone.

cs.CV↗

arXiv2022

Dual Cross-Attention Learning for Fine-Grained Visual Categorization and Object Re-Identification

Recently, self-attention mechanisms have shown impressive performance in various NLP and CV tasks, which can help capture sequential characteristics and derive global information. In this work, we explore how to extend self-attention modules to better learn subtle feature embeddings for recognizing fine-grained objects, e.g., different bird species or person identities. To this end, we propose a dual cross-attention learning (DCAL) algorithm to coordinate with self-attention learning. First, we propose global-local cross-attention (GLCA) to enhance the interactions between global images and local high-response regions, which can help reinforce the spatial-wise discriminative clues for recognition. Second, we propose pair-wise cross-attention (PWCA) to establish the interactions between image pairs. PWCA can regularize the attention learning of an image by treating another image as distractor and will be removed during inference. We observe that DCAL can reduce misleading attentions and diffuse the attention response to discover more complementary parts for recognition. We conduct extensive evaluations on fine-grained visual categorization and object re-identification. Experiments demonstrate that DCAL performs on par with state-of-the-art methods and consistently improves multiple self-attention baselines, e.g., surpassing DeiT-Tiny and ViT-Base by 2.8% and 2.4% mAP on MSMT17, respectively.

cs.CV↗

arXiv2022

Revisiting Communication-Efficient Federated Learning with Balanced Global and Local Updates

In federated learning (FL), a number of devices train their local models and upload the corresponding parameters or gradients to the base station (BS) to update the global model while protecting their data privacy. However, due to the limited computation and communication resources, the number of local trainings (a.k.a. local update) and that of aggregations (a.k.a. global update) need to be carefully chosen. In this paper, we investigate and analyze the optimal trade-off between the number of local trainings and that of global aggregations to speed up the convergence and enhance the prediction accuracy over the existing works. Our goal is to minimize the global loss function under both the delay and the energy consumption constraints. In order to make the optimization problem tractable, we derive a new and tight upper bound on the loss function, which allows us to obtain closed-form expressions for the number of local trainings and that of global aggregations. Simulation results show that our proposed scheme can achieve a better performance in terms of the prediction accuracy, and converge much faster than the baseline schemes.

cs.LG↗

arXiv2022

Nearly optimal bounds for the global geometric landscape of phase retrieval

The phase retrieval problem is concerned with recovering an unknown signal $\bf{x} \in \mathbb{R}^n$ from a set of magnitude-only measurements $y_j=|\langle \bf{a}_j,\bf{x} \rangle|, \; j=1,\ldots,m$. A natural least squares formulation can be used to solve this problem efficiently even with random initialization, despite its non-convexity of the loss function. One way to explain this surprising phenomenon is the benign geometric landscape: (1) all local minimizers are global; and (2) the objective function has a negative curvature around each saddle point and local maximizer. In this paper, we show that $m=O(n \log n)$ Gaussian random measurements are sufficient to guarantee the loss function of a commonly used estimator has such benign geometric landscape with high probability. This is a step toward answering the open problem given by Sun-Qu-Wright, in which the authors suggest that $O(n \log n)$ or even $O(n)$ is enough to guarantee the favorable geometric property.

cs.IT↗

arXiv2022

Moving structures in ultraviolet bright points: observations from Solar Orbiter/EUI

Moving structures have been detected in coronal bright points and in a solar flare in active regions, which were bi-directional, symmetrical, simultaneous, and quasi-periodic (Ning & Guo 2014; Ning 2016; Li et al. 2016a). They could be regarded as observational evidence of plasma outflows via magnetic reconnection. In this article, we explored pairs of moving structures in fifteen ultraviolet bright points (UBPs), which were observed in the quiet Sun or inside a small active region on 19 November 2020, and measured by the High Resolution (HRI) Telescopes of the Extreme Ultraviolet Imager (EUI) on board the Solar Orbiter (SolO) in two passbands, HRIEUV 174 Å and HRILyα 1216 Å. Moving structures observed in ten UBPs as starting from their bright cores and propagating toward two ends, are interpreted as diverging motions of bi-directional moving structures. These moving structures are also characterized by simultaneity and symmetry and in the case of seven UBPs they exhibit quasi-periodicity. They could be generated by outflows after magnetic reconnections. Moving structures seen in another five UBPs as originating from double ends and moving closer, and merging together, are manifested as converging motions, which might be caused by inflows through the magnetic reconnection, or might be interpreted as upflows driven by the chromospheric evaporation.

astro-ph.SR↗

arXiv2022

Compare learning: bi-attention network for few-shot learning

Learning with few labeled data is a key challenge for visual recognition, as deep neural networks tend to overfit using a few samples only. One of the Few-shot learning methods called metric learning addresses this challenge by first learning a deep distance metric to determine whether a pair of images belong to the same category, then applying the trained metric to instances from other test set with limited labels. This method makes the most of the few samples and limits the overfitting effectively. However, extant metric networks usually employ Linear classifiers or Convolutional neural networks (CNN) that are not precise enough to globally capture the subtle differences between vectors. In this paper, we propose a novel approach named Bi-attention network to compare the instances, which can measure the similarity between embeddings of instances precisely, globally and efficiently. We verify the effectiveness of our model on two benchmarks. Experiments show that our approach achieved improved accuracy and convergence speed over baseline models.

cs.CV↗

arXiv2022

Strain-dependent structural and electronic reconstructions in long-wavelength WS$_{2}$ moiré superlattices

In long-wavelength moiré superlattices of stacked transition metal dichalcogenides (TMDs), structural reconstruction ubiquitously occurs, which has reported to impact significantly their electronic properties. However, complete microscopic understandings of the interplay between the lattice reconstruction and alteration of electronic properties, and their further response to external perturbations in the reconstructed TMDs moiré superlattice are still lacking. Here, using scanning tunneling microscopy (STM) and scanning tunneling spectroscopy (STS) combined with first-principles calculation, we study the strain-dependent structural reconstruction and its correlated electronic reconstruction in long-wavelength H-type WS$_{2}$ moiré superlattice at nanometer scale. We observe that the long-wavelength WS$_{2}$ moiré superlattices experiencing strong atomic reconstruction transform into a hexagonal array of screw dislocations separating large-sized H-stacked domains. Both the geometry and the moiré wavelength of the moiré superlattice are dramatically tuned by external intralayer heterostrain in our experiment. Remarkably, the STS measurements further demonstrate that the location of the K point in conduction band is modulated sensitively by strain-induced lattice deformation at nanometer scale in this system, with the maximum energy shift reaching up to 300 meV. Our results highlight that intralayer strain plays a vital role in determining structural and electronic properties in TMD moiré superlattice.

cond-mat.mes-hall↗

arXiv2022

LHNN: Lattice Hypergraph Neural Network for VLSI Congestion Prediction

Precise congestion prediction from a placement solution plays a crucial role in circuit placement. This work proposes the lattice hypergraph (LH-graph), a novel graph formulation for circuits, which preserves netlist data during the whole learning process, and enables the congestion information propagated geometrically and topologically. Based on the formulation, we further developed a heterogeneous graph neural network architecture LHNN, jointing the routing demand regression to support the congestion spot classification. LHNN constantly achieves more than 35% improvements compared with U-nets and Pix2Pix on the F1 score. We expect our work shall highlight essential procedures using machine learning for congestion prediction.

cs.LG↗

arXiv2022

Who Should Review Your Proposal? Interdisciplinary Topic Path Detection for Research Proposals

The peer merit review of research proposals has been the major mechanism to decide grant awards. Nowadays, research proposals have become increasingly interdisciplinary. It has been a longstanding challenge to assign proposals to appropriate reviewers. One of the critical steps in reviewer assignment is to generate accurate interdisciplinary topic labels for proposals. Existing systems mainly collect topic labels manually reported by discipline investigators. However, such human-reported labels can be non-accurate and incomplete. What role can AI play in developing a fair and precise proposal review system? In this evidential study, we collaborate with the National Science Foundation of China to address the task of automated interdisciplinary topic path detection. For this purpose, we develop a deep Hierarchical Interdisciplinary Research Proposal Classification Network (HIRPCN). We first propose a hierarchical transformer to extract the textual semantic information of proposals. We then design an interdisciplinary graph and leverage GNNs to learn representations of each discipline in order to extract interdisciplinary knowledge. After extracting the semantic and interdisciplinary knowledge, we design a level-wise prediction component to fuse the two types of knowledge representations and detect interdisciplinary topic paths for each proposal. We conduct extensive experiments and expert evaluations on three real-world datasets to demonstrate the effectiveness of our proposed model.

cs.CL↗

arXiv2022

Stability and convergence of Strang splitting. Part II: tensorial Allen-Cahn equations

We consider the second-order in time Strang-splitting approximation for vector-valued and matrix-valued Allen-Cahn equations. Both the linear propagator and the nonlinear propagator are computed explicitly. For the vector-valued case, we prove the maximum principle and unconditional energy dissipation for a judiciously modified energy functional. The modified energy functional is close to the classical energy up to $\mathcal O(τ)$ where $τ$ is the splitting step. For the matrix-valued case, we prove a sharp maximum principle in the matrix Frobenius norm. We show modified energy dissipation under very mild splitting step constraints. We exhibit several numerical examples to show the efficiency of the method as well as the sharpness of the results.

math.NA↗

arXiv2022

Stability and convergence of Strang splitting. Part I: Scalar Allen-Cahn equation

We consider a class of second-order Strang splitting methods for Allen-Cahn equations with polynomial or logarithmic nonlinearities. For the polynomial case both the linear and the nonlinear propagators are computed explicitly. We show that this type of Strang splitting scheme is unconditionally stable regardless of the time step. Moreover we establish strict energy dissipation for a judiciously modified energy which coincides with the classical energy up to $\mathcal O(τ)$ where $τ$ is the time step. For the logarithmic potential case, since the continuous-time nonlinear propagator no longer enjoys explicit analytic treatments, we employ a second order in time two-stage implicit Runge--Kutta (RK) nonlinear propagator together with an efficient Newton iterative solver. We prove a maximum principle which ensures phase separation and establish energy dissipation law under mild restrictions on the time step. These appear to be the first rigorous results on the energy dissipation of Strang-type splitting methods for Allen-Cahn equations.

math.NA↗