arXiv subjects

Di Liu

Publications and source records attributed to Di Liu.

At least 109 records · Page 6Linked to original sources

arXiv2023

A successive centralized circumcenter reflection method for the convex feasibility problem

In this paper we present the successive centralization of the circumcenter reflection scheme with several control sequences for solving the convex feasibility problem in Euclidean space. Assuming that a standard error bound holds, we prove the linear convergence of the method with the most violated constraint control sequence. Under additional smoothness assumptions, we prove the superlinear convergence. Numerical experiments confirm the efficiency of our method.

math.OC↗

arXiv2023

Complete mode conversion for elastic waves reflected by elastic metamaterial slab with double hexapole resonances

In this study, we investigate the phenomenon of mode conversion in elastic bulk waves using coupled hexapole resonances. A metamaterial slab is proposed enabling the complete conversion between longitudinal and transverse modes. Each unit of the elastic metamaterial slab comprises a pair of scatterers, and their relative direction is oriented at an oblique angle. The interaction between the coupled hexapoles and the background results in oblique displacements, which are responsible for the mode conversion. Moreover, this conversion exhibits a broader frequency range compared to the quadrupole resonance. This innovative design significantly broadens the range of possibilities for developing mode-converting metamaterials.

physics.app-ph↗

arXiv2023

The silicon vacancy centers in SiC: determination of intrinsic spin dynamics for integrated quantum photonics

The negatively-charged silicon vacancy center ($\rm V_{Si}^-$) in silicon carbide (SiC) is an emerging color center for quantum technology covering quantum sensing, communication, and computing. Yet, limited information currently available on the internal spin-optical dynamics of these color centers prevents us achieving the optimal operation conditions and reaching the maximum performance especially when integrated within quantum photonics. Here, we establish all the relevant intrinsic spin dynamics of negatively charged $\rm V_{Si}^-$ center in 4H-SiC by an in-depth electronic fine structure modeling including intersystem-crossing and deshelving mechanisms. With carefully designed spin-dependent measurements, we obtain all previously unknown spin-selective radiative and non-radiative decay rates. To showcase the relevance of our work for integrated quantum photonics, we use the obtained rates to propose a realistic implementation of time-bin entangled multi-photon GHZ and cluster state generation. We find that up to 3-photon GHZ/cluster states are readily within reach using the existing nanophotonic cavity technology.

quant-ph↗

arXiv2023

Improving Tuning-Free Real Image Editing with Proximal Guidance

DDIM inversion has revealed the remarkable potential of real image editing within diffusion-based methods. However, the accuracy of DDIM reconstruction degrades as larger classifier-free guidance (CFG) scales being used for enhanced editing. Null-text inversion (NTI) optimizes null embeddings to align the reconstruction and inversion trajectories with larger CFG scales, enabling real image editing with cross-attention control. Negative-prompt inversion (NPI) further offers a training-free closed-form solution of NTI. However, it may introduce artifacts and is still constrained by DDIM reconstruction quality. To overcome these limitations, we propose proximal guidance and incorporate it to NPI with cross-attention control. We enhance NPI with a regularization term and reconstruction guidance, which reduces artifacts while capitalizing on its training-free nature. Additionally, we extend the concepts to incorporate mutual self-attention control, enabling geometry and layout alterations in the editing process. Our method provides an efficient and straightforward approach, effectively addressing real image editing tasks with minimal computational overhead.

cs.CV↗

arXiv2023

TransDetector: A Transformer-Based Detector for Underwater Acoustic Differential OFDM Communications

Inter-carrier interference (ICI) and noise mitigation is crucial for precise signal detection in underwater acoustic (UWA) differential orthogonal frequency division multiplexing (DOFDM) communication systems. In this paper, we adopt the Transformer to design a detector, referred to as the TransDetector, which can dramatically mitigate ICI implicitly and noise explicitly, even without requiring any pilot. Compared with the standard Transformer, we come up with three creative designs. Firstly, we break the inner-encoder computation paradigm of the multi-head attention (MHA) in the standard Transformer, and design a brand new inter-encoder attention mechanism, referred to as the interactive MHA, which can significantly improve the performance, as well as accelerate the convergence rate. Secondly, to reduce the noise component attached to the received signal, we design an auto-perception denoising structure, which allows the network to learn the noise distribution in received signals. Thirdly, to better match the characteristics of DOFDM signals and selectively focus on the data at specified locations, we propose a trapezoidal positional encoding (PE), instead of adopting the original sine-cosine PE in the Transformer. Experimental results on both the realistic underwater channel and the simulation channel show that the TransDetector outperforms the classical $\mathcal{X}$-FFT algorithms and the DNNDetector in terms of the BER and the MSE. For example, the BER achieved by the TransDetector is reduced by $27.21\%$ and $12.50\%$ when the signal-to-noise ratio $(\text{SNR})=0$~dB and by $47.44\%$ and $33.49\%$ when $\text{SNR}=20$~dB against the PS-FFT and the DNNDetector based on the realistic channel, respectively.

cs.IT↗

arXiv2023

Cooperative IoT Data Sharing with Heterogeneity of Participants Based on Electricity Retail

With the development of Internet of Things (IoT) and big data technology, the data value is increasingly explored in multiple practical scenarios, including electricity transactions. However, the isolation of IoT data among several entities makes it difficult to achieve optimal allocation of data resources and convert data resources into real economic value, thus it is necessary to introduce the IoT data sharing mode to drive data circulation. To enhance the accuracy and fairness of IoT data sharing, the heterogeneity of participants is sufficiently considered, and data valuation and profit allocation in IoT data sharing are improved based on the background of electricity retail. Data valuation is supposed to be relevant to attributes of IoT data buyers, thus risk preferences of electricity retailers are applied as characteristic attributes and data premium rates are proposed to modify data value rates. Profit allocation should measure the marginal contribution shares of electricity retailers and data brokers fairly, thus asymmetric Nash bargaining model is used to guarantee that they could receive reasonable profits based on their specific contribution to the coalition of IoT data sharing. Considering the heterogeneity of participants comprehensively, the proposed IoT data sharing fits for a large coalition of IoT data sharing with multiple electricity retailers and data brokers. Finally, to demonstrate the applications of IoT data sharing in smart grids, case studies are utilized to validate the results of data value for electricity retailers with different risk preferences and the efficiency of profit allocation using asymmetric Nash bargaining model.

eess.SY↗

arXiv2023

Discovery of two rotational modulation periods from a young hierarchical triple system

GW~Ori is a young hierarchical triple system located in $λ$ Orionis, consisting of a binary (GW~Ori\,A and B), a tertiary star (GW~Ori\,C) and a rare circumtriple disk. Due to the limited data with poor accuracy, several short-period signals were detected in this system, but the values from different studies are not fully consistent. As one of the most successful transiting surveys, the Transiting Exoplanet Survey Satellite (TESS) provides an unprecedented opportunity to make a comprehensive periodic analysis of GW~Ori. In this work we discover two significant modulation signals by analyzing the light curves of GW~Ori's four observations from TESS, i.e., 3.02 $\pm$ 0.15\,d and 1.92 $\pm$ 0.06\,d, which are very likely to be the rotational periods caused by starspot modulation on the primary and secondary components, respectively. We calculate the inclinations of GW~Ori\,A and B according to the two rotational periods. The results suggest that the rotational plane of GW~Ori\,A and B and the orbital plane of the binary are almost coplanar. We also discuss the aperiodic features in the light curves; these may be related to unstable accretion. The light curves of GW~Ori also include a third (possible) modulation signal with a period of 2.51$\pm$0.09\,d, but the third is neither quite stable nor statistically significant.

astro-ph.SR↗

arXiv2023

Dealing With Heterogeneous 3D MR Knee Images: A Federated Few-Shot Learning Method With Dual Knowledge Distillation

Federated Learning has gained popularity among medical institutions since it enables collaborative training between clients (e.g., hospitals) without aggregating data. However, due to the high cost associated with creating annotations, especially for large 3D image datasets, clinical institutions do not have enough supervised data for training locally. Thus, the performance of the collaborative model is subpar under limited supervision. On the other hand, large institutions have the resources to compile data repositories with high-resolution images and labels. Therefore, individual clients can utilize the knowledge acquired in the public data repositories to mitigate the shortage of private annotated images. In this paper, we propose a federated few-shot learning method with dual knowledge distillation. This method allows joint training with limited annotations across clients without jeopardizing privacy. The supervised learning of the proposed method extracts features from limited labeled data in each client, while the unsupervised data is used to distill both feature and response-based knowledge from a national data repository to further improve the accuracy of the collaborative model and reduce the communication cost. Extensive evaluations are conducted on 3D magnetic resonance knee images from a private clinical dataset. Our proposed method shows superior performance and less training time than other semi-supervised federated learning methods. Codes and additional visualization results are available at https://github.com/hexiaoxiao-cs/fedml-knee.

eess.IV↗

arXiv2023

A Data-scalable Transformer for Medical Image Segmentation: Architecture, Model Efficiency, and Benchmark

Transformers have demonstrated remarkable performance in natural language processing and computer vision. However, existing vision Transformers struggle to learn from limited medical data and are unable to generalize on diverse medical image tasks. To tackle these challenges, we present MedFormer, a data-scalable Transformer designed for generalizable 3D medical image segmentation. Our approach incorporates three key elements: a desirable inductive bias, hierarchical modeling with linear-complexity attention, and multi-scale feature fusion that integrates spatial and semantic information globally. MedFormer can learn across tiny- to large-scale data without pre-training. Comprehensive experiments demonstrate MedFormer's potential as a versatile segmentation backbone, outperforming CNNs and vision Transformers on seven public datasets covering multiple modalities (e.g., CT and MRI) and various medical targets (e.g., healthy organs, diseased tissues, and tumors). We provide public access to our models and evaluation pipeline, offering solid baselines and unbiased comparisons to advance a wide range of downstream clinical applications.

eess.IV↗

arXiv2023

Improving Generalization with Domain Convex Game

Domain generalization (DG) tends to alleviate the poor generalization capability of deep neural networks by learning model with multiple source domains. A classical solution to DG is domain augmentation, the common belief of which is that diversifying source domains will be conducive to the out-of-distribution generalization. However, these claims are understood intuitively, rather than mathematically. Our explorations empirically reveal that the correlation between model generalization and the diversity of domains may be not strictly positive, which limits the effectiveness of domain augmentation. This work therefore aim to guarantee and further enhance the validity of this strand. To this end, we propose a new perspective on DG that recasts it as a convex game between domains. We first encourage each diversified domain to enhance model generalization by elaborately designing a regularization term based on supermodularity. Meanwhile, a sample filter is constructed to eliminate low-quality samples, thereby avoiding the impact of potentially harmful information. Our framework presents a new avenue for the formal analysis of DG, heuristic analysis and extensive experiments demonstrate the rationality and effectiveness.

cs.CV↗

arXiv2023

On-chip quantum interference between the origins of a multi-photon state

Path identiy induces a broad interest in recent years due to the foundation for numerous novel quantum information applications. Here, we experimentally demonstrate quantum coherent superposition of two different origins of a four-photon state, where multi-photon frustrated interference emerges from the quantum indistinguishability by path identity. The quantum state is created in four probabilistic photon-pair sources on one integrated silicon photonic chip, two combinations of which can create photon quadruplets. Coherent elimination and revival of distributed four-photons are fully controlled by tuning phases. The experiment gives rise to peculiar quantum interference of two possible ways to create photon quadruplets rather than interference of different intrinsic properties of photons. Besides many known potential applications, this new kind of multi-photon nonlinear interference enables the possibility for various fundamental studies such as nonlocality with multiple spatially separated locations.

quant-ph↗

arXiv2022

Multi-scale frequency separation network for image deblurring

Image deblurring aims to restore the detailed texture information or structures from blurry images, which has become an indispensable step in many computer vision tasks. Although various methods have been proposed to deal with the image deblurring problem, most of them treated the blurry image as a whole and neglected the characteristics of different image frequencies. In this paper, we present a new method called multi-scale frequency separation network (MSFS-Net) for image deblurring. MSFS-Net introduces the frequency separation module (FSM) into an encoder-decoder network architecture to capture the low- and high-frequency information of image at multiple scales. Then, a cycle-consistency strategy and a contrastive learning module (CLM) are respectively designed to retain the low-frequency information and recover the high-frequency information during deblurring. At last, the features of different scales are fused by a cross-scale feature fusion module (CSFFM). Extensive experiments on benchmark datasets show that the proposed network achieves state-of-the-art performance.

cs.CV↗

arXiv2022

Region Proposal Rectification Towards Robust Instance Segmentation of Biological Images

Top-down instance segmentation framework has shown its superiority in object detection compared to the bottom-up framework. While it is efficient in addressing over-segmentation, top-down instance segmentation suffers from over-crop problem. However, a complete segmentation mask is crucial for biological image analysis as it delivers important morphological properties such as shapes and volumes. In this paper, we propose a region proposal rectification (RPR) module to address this challenging incomplete segmentation problem. In particular, we offer a progressive ROIAlign module to introduce neighbor information into a series of ROIs gradually. The ROI features are fed into an attentive feed-forward network (FFN) for proposal box regression. With additional neighbor information, the proposed RPR module shows significant improvement in correction of region proposal locations and thereby exhibits favorable instance segmentation performances on three biological image datasets compared to state-of-the-art baseline methods. Experimental results demonstrate that the proposed RPR module is effective in both anchor-based and anchor-free top-down instance segmentation approaches, suggesting the proposed method can be applied to general top-down instance segmentation of biological images.

cs.CV↗

arXiv2022

TransFusion: Multi-view Divergent Fusion for Medical Image Segmentation with Transformers

Combining information from multi-view images is crucial to improve the performance and robustness of automated methods for disease diagnosis. However, due to the non-alignment characteristics of multi-view images, building correlation and data fusion across views largely remain an open problem. In this study, we present TransFusion, a Transformer-based architecture to merge divergent multi-view imaging information using convolutional layers and powerful attention mechanisms. In particular, the Divergent Fusion Attention (DiFA) module is proposed for rich cross-view context modeling and semantic dependency mining, addressing the critical issue of capturing long-range correlations between unaligned data from different image views. We further propose the Multi-Scale Attention (MSA) to collect global correspondence of multi-scale feature representations. We evaluate TransFusion on the Multi-Disease, Multi-View \& Multi-Center Right Ventricular Segmentation in Cardiac MRI (M\&Ms-2) challenge cohort. TransFusion demonstrates leading performance against the state-of-the-art methods and opens up new perspectives for multi-view imaging integration towards robust medical image segmentation.

eess.IV↗

arXiv2022

You Only Search Once: On Lightweight Differentiable Architecture Search for Resource-Constrained Embedded Platforms

Benefiting from the search efficiency, differentiable neural architecture search (NAS) has evolved as the most dominant alternative to automatically design competitive deep neural networks (DNNs). We note that DNNs must be executed under strictly hard performance constraints in real-world scenarios, for example, the runtime latency on autonomous vehicles. However, to obtain the architecture that meets the given performance constraint, previous hardware-aware differentiable NAS methods have to repeat a plethora of search runs to manually tune the hyper-parameters by trial and error, and thus the total design cost increases proportionally. To resolve this, we introduce a lightweight hardware-aware differentiable NAS framework dubbed LightNAS, striving to find the required architecture that satisfies various performance constraints through a one-time search (i.e., \underline{\textit{you only search once}}). Extensive experiments are conducted to show the superiority of LightNAS over previous state-of-the-art methods.

cs.LG↗

arXiv2022

FAT: An In-Memory Accelerator with Fast Addition for Ternary Weight Neural Networks

Convolutional Neural Networks (CNNs) demonstrate excellent performance in various applications but have high computational complexity. Quantization is applied to reduce the latency and storage cost of CNNs. Among the quantization methods, Binary and Ternary Weight Networks (BWNs and TWNs) have a unique advantage over 8-bit and 4-bit quantization. They replace the multiplication operations in CNNs with additions, which are favoured on In-Memory-Computing (IMC) devices. IMC acceleration for BWNs has been widely studied. However, though TWNs have higher accuracy and better sparsity than BWNs, IMC acceleration for TWNs has limited research. TWNs on existing IMC devices are inefficient because the sparsity is not well utilized, and the addition operation is not efficient. In this paper, we propose FAT as a novel IMC accelerator for TWNs. First, we propose a Sparse Addition Control Unit, which utilizes the sparsity of TWNs to skip the null operations on zero weights. Second, we propose a fast addition scheme based on the memory Sense Amplifier to avoid the time overhead of both carry propagation and writing back the carry to memory cells. Third, we further propose a Combined-Stationary data mapping to reduce the data movement of activations and weights and increase the parallelism across memory columns. Simulation results show that for addition operations at the Sense Amplifier level, FAT achieves 2.00X speedup, 1.22X power efficiency, and 1.22X area efficiency compared with a State-Of-The-Art IMC accelerator ParaPIM. FAT achieves 10.02X speedup and 12.19X energy efficiency compared with ParaPIM on networks with 80% average sparsity.

cs.AR↗

arXiv2022

DeepRecon: Joint 2D Cardiac Segmentation and 3D Volume Reconstruction via A Structure-Specific Generative Method

Joint 2D cardiac segmentation and 3D volume reconstruction are fundamental to building statistical cardiac anatomy models and understanding functional mechanisms from motion patterns. However, due to the low through-plane resolution of cine MR and high inter-subject variance, accurately segmenting cardiac images and reconstructing the 3D volume are challenging. In this study, we propose an end-to-end latent-space-based framework, DeepRecon, that generates multiple clinically essential outcomes, including accurate image segmentation, synthetic high-resolution 3D image, and 3D reconstructed volume. Our method identifies the optimal latent representation of the cine image that contains accurate semantic information for cardiac structures. In particular, our model jointly generates synthetic images with accurate semantic information and segmentation of the cardiac structures using the optimal latent representation. We further explore downstream applications of 3D shape reconstruction and 4D motion pattern adaptation by the different latent-space manipulation strategies.The simultaneously generated high-resolution images present a high interpretable value to assess the cardiac shape and motion.Experimental results demonstrate the effectiveness of our approach on multiple fronts including 2D segmentation, 3D reconstruction, downstream 4D motion pattern adaption performance.

cs.CV↗

arXiv2022

Leading Two-loop corrections to the mass of Higgs boson in the High scale Dirac gaugino supersymmetry

Precision measurements of the Higgs mass have become a powerful constraint on models of physics beyond the standard model. We revisit supersymmetric models with Dirac gauginos and study the contributions to the Higgs mass. We calculate the leading two-loop corrections to the SM-like Higgs mass by constructing a series of EFTs and iteratively integrating out heavy particles. We then apply these calculations to a variety of scenarios, including a simple Dirac gluino, and split Dirac models of supersymmetry. We present the detailed formulae for threshold corrections and compare with previous results, where available. In general, the contributions are small, but the additional precision allows us to make more concrete statements about the relevant scales in Dirac SUSY models.

hep-ph↗