arXiv subjects

Min Li

Publications and source records attributed to Min Li.

At least 181 records · Page 10Linked to original sources

arXiv2023

INGB: Informed Nonlinear Granular Ball Oversampling Framework for Noisy Imbalanced Classification

In classification problems, the datasets are usually imbalanced, noisy or complex. Most sampling algorithms only make some improvements to the linear sampling mechanism of the synthetic minority oversampling technique (SMOTE). Nevertheless, linear oversampling has several unavoidable drawbacks. Linear oversampling is susceptible to overfitting, and the synthetic samples lack diversity and rarely account for the original distribution characteristics. An informed nonlinear oversampling framework with the granular ball (INGB) as a new direction of oversampling is proposed in this paper. It uses granular balls to simulate the spatial distribution characteristics of datasets, and informed entropy is utilized to further optimize the granular-ball space. Then, nonlinear oversampling is performed by following high-dimensional sparsity and the isotropic Gaussian distribution. Furthermore, INGB has good compatibility. Not only can it be combined with most SMOTE-based sampling algorithms to improve their performance, but it can also be easily extended to noisy imbalanced multi-classification problems. The mathematical model and theoretical proof of INGB are given in this work. Extensive experiments demonstrate that INGB outperforms the traditional linear sampling frameworks and algorithms in oversampling on complex datasets.

stat.ML↗

arXiv2023

Multimodal Zero-Shot Learning for Tactile Texture Recognition

Tactile sensing plays an irreplaceable role in robotic material recognition. It enables robots to distinguish material properties such as their local geometry and textures, especially for materials like textiles. However, most tactile recognition methods can only classify known materials that have been touched and trained with tactile data, yet cannot classify unknown materials that are not trained with tactile data. To solve this problem, we propose a tactile zero-shot learning framework to recognise unknown materials when they are touched for the first time without requiring training tactile samples. The visual modality, providing tactile cues from sight, and semantic attributes, giving high-level characteristics, are combined together to bridge the gap between touched classes and untouched classes. A generative model is learnt to synthesise tactile features according to corresponding visual images and semantic embeddings, and then a classifier can be trained using the synthesised tactile features of untouched materials for zero-shot recognition. Extensive experiments demonstrate that our proposed multimodal generative model can achieve a high recognition accuracy of 83.06% in classifying materials that were not touched before. The robotic experiment demo and the dataset are available at https://sites.google.com/view/multimodalzsl.

cs.RO↗

arXiv2023

Fast Segment Anything

The recently proposed segment anything model (SAM) has made a significant influence in many computer vision tasks. It is becoming a foundation step for many high-level tasks, like image segmentation, image caption, and image editing. However, its huge computation costs prevent it from wider applications in industry scenarios. The computation mainly comes from the Transformer architecture at high-resolution inputs. In this paper, we propose a speed-up alternative method for this fundamental task with comparable performance. By reformulating the task as segments-generation and prompting, we find that a regular CNN detector with an instance segmentation branch can also accomplish this task well. Specifically, we convert this task to the well-studied instance segmentation task and directly train the existing instance segmentation method using only 1/50 of the SA-1B dataset published by SAM authors. With our method, we achieve a comparable performance with the SAM method at 50 times higher run-time speed. We give sufficient experimental results to demonstrate its effectiveness. The codes and demos will be released at https://github.com/CASIA-IVA-Lab/FastSAM.

cs.CV↗

arXiv2023

DeepGate2: Functionality-Aware Circuit Representation Learning

Circuit representation learning aims to obtain neural representations of circuit elements and has emerged as a promising research direction that can be applied to various EDA and logic reasoning tasks. Existing solutions, such as DeepGate, have the potential to embed both circuit structural information and functional behavior. However, their capabilities are limited due to weak supervision or flawed model design, resulting in unsatisfactory performance in downstream tasks. In this paper, we introduce DeepGate2, a novel functionality-aware learning framework that significantly improves upon the original DeepGate solution in terms of both learning effectiveness and efficiency. Our approach involves using pairwise truth table differences between sampled logic gates as training supervision, along with a well-designed and scalable loss function that explicitly considers circuit functionality. Additionally, we consider inherent circuit characteristics and design an efficient one-round graph neural network (GNN), resulting in an order of magnitude faster learning speed than the original DeepGate solution. Experimental results demonstrate significant improvements in two practical downstream tasks: logic synthesis and Boolean satisfiability solving. The code is available at https://github.com/cure-lab/DeepGate2

cs.LG↗

arXiv2023

Multivariate Intrinsic Local Polynomial Regression on Isometric Riemannian Manifolds: Applications to Positive Definite Data

The paper introduces a novel non-parametric Riemannian regression method using Isometric Riemannian Manifolds (IRMs). The proposed technique, Intrinsic Local Polynomial Regression on IRMs (ILPR-IRMs), enables global data mapping between Riemannian manifolds while preserving underlying geometries. The ILPR method is generalized to handle multivariate covariates on any Riemannian manifold and isometry. Specifically, for manifolds equipped with Euclidean Pullback Metrics (EPMs), a closed analytical formula is derived for the multivariate ILPR (ILPR-EPM). Asymptotic statistical properties of the ILPR-EPM for the multivariate local linear case are established, including a formula for the asymptotic bias, establishing estimator consistency. The paper showcases possible applications of the method by focusing on a group of Riemannian metrics on the Symmetric Positive Definite (SPD) manifold, which arises in machine learning and neuroscience. It is demonstrated that several metrics on the SPD manifold are EPMs, resulting in a closed analytical expression for the multivariate ILPR estimator on the SPD manifold. The paper evaluates the ILPR estimator's performance under two specific EPMs, Log-Cholesky and Log-Euclidean, on simulated data on the SPD manifold and compares it with extrinsic LPR using the Affine-Invariant when scaling the manifold and covariate dimension. The results show that the ILPR using the Log-Cholesky metric is computationally faster and provides a better trade-off between error and time than other metrics. Finally, the Log-Cholesky metric on the SPD manifold is employed to implement an efficient and intrinsic version of Rie-SNE for visualizing high-dimensional SPD data. The code for implementing ILPR-EPMs and other relevant calculations is available on the GitHub page.

stat.ME↗

arXiv2023

Addressing Variable Dependency in GNN-based SAT Solving

Boolean satisfiability problem (SAT) is fundamental to many applications. Existing works have used graph neural networks (GNNs) for (approximate) SAT solving. Typical GNN-based end-to-end SAT solvers predict SAT solutions concurrently. We show that for a group of symmetric SAT problems, the concurrent prediction is guaranteed to produce a wrong answer because it neglects the dependency among Boolean variables in SAT problems. % We propose AsymSAT, a GNN-based architecture which integrates recurrent neural networks to generate dependent predictions for variable assignments. The experiment results show that dependent variable prediction extends the solving capability of the GNN-based method as it improves the number of solved SAT instances on large test sets.

cs.AI↗

arXiv2023

Peak-First CTC: Reducing the Peak Latency of CTC Models by Applying Peak-First Regularization

The CTC model has been widely applied to many application scenarios because of its simple structure, excellent performance, and fast inference speed. There are many peaks in the probability distribution predicted by the CTC models, and each peak represents a non-blank token. The recognition latency of CTC models can be reduced by encouraging the model to predict peaks earlier. Existing methods to reduce latency require modifying the transition relationship between tokens in the forward-backward algorithm, and the gradient calculation. Some of these methods even depend on the forced alignment results provided by other pretrained models. The above methods are complex to implement. To reduce the peak latency, we propose a simple and novel method named peak-first regularization, which utilizes a frame-wise knowledge distillation function to force the probability distribution of the CTC model to shift left along the time axis instead of directly modifying the calculation process of CTC loss and gradients. All the experiments are conducted on a Chinese Mandarin dataset AISHELL-1. We have verified the effectiveness of the proposed regularization on both streaming and non-streaming CTC models respectively. The results show that the proposed method can reduce the average peak latency by about 100 to 200 milliseconds with almost no degradation of recognition accuracy.

eess.AS↗

arXiv2023

First sodium laser guide star asterism launching platform in China on 1.8m telescope at Gaomeigu Observatory

The application of sodium laser guide star is the key difference between modern adaptive optics system and traditional adaptive optics system. Especially in system like multi-conjugate adaptive optics, sodium laser guide star asterism which is formed by several laser guide stars in certain pattern is required to probe more atmospheric turbulence in different directions. To achieve this, a sodium laser guide star asterism launching platform is required. In this paper, we will introduce the sodium laser guide star asterism launching platform built and tested on the 1.8m telescope of the Gaomeigu Observatory. The platform has two functions: one is to compare the performance of sodium laser guide stars generated by different lasers at the same place; the other is to generate sodium laser guide star asterism with adjustable shape. The field test results at the beginning of 2021 verify the important role of the platform, which is also the first time to realize sodium laser guide star asterism in China.

astro-ph.IM↗

arXiv2023

TA-MoE: Topology-Aware Large Scale Mixture-of-Expert Training

Sparsely gated Mixture-of-Expert (MoE) has demonstrated its effectiveness in scaling up deep neural networks to an extreme scale. Despite that numerous efforts have been made to improve the performance of MoE from the model design or system optimization perspective, existing MoE dispatch patterns are still not able to fully exploit the underlying heterogeneous network environments. In this paper, we propose TA-MoE, a topology-aware routing strategy for large-scale MoE trainging, from a model-system co-design perspective, which can dynamically adjust the MoE dispatch pattern according to the network topology. Based on communication modeling, we abstract the dispatch problem into an optimization objective and obtain the approximate dispatch pattern under different topologies. On top of that, we design a topology-aware auxiliary loss, which can adaptively route the data to fit in the underlying topology without sacrificing the model accuracy. Experiments show that TA-MoE can substantially outperform its counterparts on various hardware and model configurations, with roughly 1.01x-1.61x, 1.01x-4.77x, 1.25x-1.54x improvements over the popular DeepSpeed-MoE, FastMoE and FasterMoE.

cs.LG↗

arXiv2023

Ill-posedness for a generalized Camassa-Holm equation with higher-order nonlinearity in the critical Besov space

In this paper, we prove that the Cauchy problem for a generalized Camassa-Holm equation with higher-order nonlinearity is ill-posed in the critical Besov space $B^1_{\infty,1}(\R)$. It is shown in (J. Differ. Equ., 327:127-144,2022) that the Camassa-Holm equation is ill-posed in $B^1_{\infty,1}(\R)$, here we turn our attention to a higher-order nonlinear generalization of Camassa-Holm equation proposed by Hakkaev and Kirchev (Commun Partial Differ Equ 30:761-781,2005). With newly constructed initial data, we get the norm inflation in the critical space $B^1_{\infty,1}(\R)$ which leads to ill-posedness.

math.AP↗

arXiv2023

DeepSAT: An EDA-Driven Learning Framework for SAT

We present DeepSAT, a novel end-to-end learning framework for the Boolean satisfiability (SAT) problem. Unlike existing solutions trained on random SAT instances with relatively weak supervision, we propose applying the knowledge of the well-developed electronic design automation (EDA) field for SAT solving. Specifically, we first resort to logic synthesis algorithms to pre-process SAT instances into optimized and-inverter graphs (AIGs). By doing so, the distribution diversity among various SAT instances can be dramatically reduced, which facilitates improving the generalization capability of the learned model. Next, we regard the distribution of SAT solutions being a product of conditional Bernoulli distributions. Based on this observation, we approximate the SAT solving procedure with a conditional generative model, leveraging a novel directed acyclic graph neural network (DAGNN) with two polarity prototypes for conditional SAT modeling. To effectively train the generative model, with the help of logic simulation tools, we obtain the probabilities of nodes in the AIG being logic `1' as rich supervision. We conduct comprehensive experiments on various SAT problems. Our results show that, DeepSAT achieves significant accuracy improvements over state-of-the-art learning-based SAT solutions, especially when generalized to SAT instances that are relatively large or with diverse distributions.

cs.AI↗

arXiv2023

Vis2Hap: Vision-based Haptic Rendering by Cross-modal Generation

To assist robots in teleoperation tasks, haptic rendering which allows human operators access a virtual touch feeling has been developed in recent years. Most previous haptic rendering methods strongly rely on data collected by tactile sensors. However, tactile data is not widely available for robots due to their limited reachable space and the restrictions of tactile sensors. To eliminate the need for tactile data, in this paper we propose a novel method named as Vis2Hap to generate haptic rendering from visual inputs that can be obtained from a distance without physical interaction. We take the surface texture of objects as key cues to be conveyed to the human operator. To this end, a generative model is designed to simulate the roughness and slipperiness of the object's surface. To embed haptic cues in Vis2Hap, we use height maps from tactile sensors and spectrograms from friction coefficients as the intermediate outputs of the generative model. Once Vis2Hap is trained, it can be used to generate height maps and spectrograms of new surface textures, from which a friction image can be obtained and displayed on a haptic display. The user study demonstrates that our proposed Vis2Hap method enables users to access a realistic haptic feeling similar to that of physical objects. The proposed vision-based haptic rendering has the potential to enhance human operators' perception of the remote environment and facilitate robotic manipulation.

cs.RO↗

arXiv2023

Imaging of muon track in CsI(Tl) crystal with single photon sensitive camera

As a novel approach on visual photon imaging by a single photon sensitive camera and PMTs, this work is trying to measure and identify muon tracks from the 2-D images of CsI(Tl) crystal (scintillator detectors). It is possible that muon tracks can be seen directly with a good signal-to-noise ratio neither with further amplification nor external light, which provides an evolution method for particle measurement in the photon-starved regime of scintillation detectors. The setup of the crystal and camera testing system and the identification algorithm of muon track will be discussed in detail including the system calibration, identification model, signal-to-noise ratio, muon track confirmation, and an expectation on further improvements and applications.

physics.ins-det↗

arXiv2022

Study on the linearity of 20" dynode and MCP PMTs

The linearity of charge response is an important feature of photomultiplier tubes (PMT) for physics measurements, especially the newly developed 20" MCP-PMT. In this paper, in addition to the traditional method of double light sources, we applied another relative method of 20" PMT to 3" PMT to measure the linearity of the 20" dynode and MCP PMTs in pulse mode with a waveform digitizer. The measurements show a good linear response of 20" PMTs to 1,000 photoelectrons (p.e.). The correlations of the amplitude, rise-time, fall-time, FWHM, baseline recovery, overshoot, and late-pulse to the output charge of the 20" PMTs derived from the waveform analysis, where the MCP-PMT shows very different features compared to the dynode-PMT in particular.

physics.ins-det↗

arXiv2022

Asymptotically isometric codes for holography

The holographic principle suggests that the low energy effective field theory of gravity, as used to describe perturbative quantum fields about some background has far too many states. It is then natural that any quantum error correcting code with such a quantum field theory as the code subspace is not isometric. We discuss how this framework can naturally arise in an algebraic QFT treatment of a family of CFT with a large-$N$ limit described by the single trace sector. We show that an isometric code can be recovered in the $N \rightarrow \infty$ limit when acting on fixed states in the code Hilbert space. Asymptotically isometric codes come equipped with the notion of simple operators and nets of causal wedges. While the causal wedges are additive, they need not satisfy Haag duality, thus leading to the possibility of non-trivial entanglement wedge reconstructions. Codes with complementary recovery are defined as having extensions to Haag dual nets, where entanglement wedges are well defined for all causal boundary regions. We prove an asymptotic version of the information disturbance trade-off theorem and use this to show that boundary theory causality is maintained by net extensions. We give a characterization of the existence of an entanglement wedge extension via the asymptotic equality of bulk and boundary relative entropy or modular flow. While these codes are asymptotically exact, at fixed $N$ they can have large errors on states that do not survive the large-$N$ limit. This allows us to fix well known issues that arise when modeling gravity as an exact codes, while maintaining the nice features expected of gravity, including, among other things, the emergence of non-trivial von Neumann algebras of various types.

hep-th↗

arXiv2022

On the continuity of the solution map of the Euler-Poincaré equations in Besov spaces

By constructing a series of perturbation functions through localization in the Fourier domain and translation, we show that the data-to-solution map for the Euler-Poincaré equations is nowhere uniformly continuous in $B^s_{p,r}(\mathbb{R} ^d)$ with $s>\max\{1+\frac d2,\frac32\}$ and $(p,r)\in (1,\infty)\times [1,\infty)$. This improves our previous result which shows the data-to-solution map for the Euler-Poincaré equations is non-uniformly continuous on a bounded subset of $B^s_{p,r}(\mathbb{R} ^d)$ near the origin.

math.AP↗

arXiv2022

Magnetic-Field effect in Laser-assisted XUV ionization

The magnetic-field effect of the laser pulse is investigated in laser-assisted XUV ionization. By numerically solving the three-dimensional time-dependent Schrödinger equation, we find that the photoelectron momentum distribution is distorted by the magnetic-field effect of the IR streaking field. It results in a transverse-momentum- and time-delay-dependent longitudinal photoelectron momentum shift, which is further analytically confirmed by employing the nondipole corrected strong field approximation. We also demonstrate that the magnetic field does not affect the time shift retrieved from the streaking spectrum, although the streaking spectrum is apparently altered by the momentum shift induced by the magnetic-field effect. Our work reveals the time-resolved nondipole effect of the laser pulse by the attosecond streaking technique.

physics.atom-ph↗

arXiv2022

Identification of oscillatory brain networks with Hidden Gaussian Graphical Spectral models of EEG/MEG

Identifying the functional networks underpinning indirectly observed processes poses an inverse problem for neurosciences or other fields. A solution of such inverse problems estimates as a first step the activity emerging within functional networks from EEG or MEG data. These EEG or MEG estimates are a direct reflect functional brain network activity with a temporal resolution that no other in vivo neuroimage may provide. A second step estimating functional connectivity from such activity pseudodata unveil the oscillatory brain networks that strongly correlate with all cognition and behavior. Simulations of such MEG or EEG inverse problem also reveal estimation errors of the functional connectivity determined by any of the state-of-the-art inverse solutions. We disclose a significant cause of estimation errors originating from misspecification of the functional network model incorporated into either inverse solution steps. We introduce the Bayesian identification of a Hidden Gaussian Graphical Spectral (HIGGS) model specifying such oscillatory brain networks model. In human EEG alpha rhythm simulations estimation errors measured as ROC performance do not surpass 2% in our HIGGS inverse solution and reach 20% in state-of-the-art methods. Macaque simultaneous EEG/ECoG recordings provide experimental confirmation for our inverse-solution with 1/3 more congruence according to Riemannian distances than state-of-the-art methods.

stat.ME↗