Peng Wei, Sujit Manna, Marius Eich, Patrick Lee, Jagadeesh Moodera
The induced superconductivity (SC) in a robust and scalable quantum material with strong Rashba spin-orbit coupling is particularly attractive for generating topological superconductivity and Majorana bound states (MBS). Gold (111) thin film has been proposed as a promising candidate because of the large Rashba energy, the predicted topological nature and the possibility for large-scale MBS device fabrications. We experimentally demonstrate two important steps towards achieving such a goal. We successfully show induced SC in the Shockley surface state (SS) of ultrathin Au(111) layers grown over epitaxial vanadium films, which is easily achievable on a wafer scale. The emergence of SC in the SS, which is physically separated from a bulk superconductor, is attained by indirect quasiparticle scattering processes instead of by conventional interfacial Andreev reflections. We further show the ability to tune the SS Fermi level (E_F) by interfacing SS with a high-k dielectric ferromagnetic insulator EuS. The shift of E_F from ~ 550 mV to ~34mV in superconducting SS is an important step towards realizing MBS in this robust system.
Jason Cong, Peng Wei, Cody Hao Yu, Peng Zhang
CPU-FPGA heterogeneous architectures are attracting ever-increasing attention in an attempt to advance computational capabilities and energy efficiency in today's datacenters. These architectures provide programmers with the ability to reprogram the FPGAs for flexible acceleration of many workloads. Nonetheless, this advantage is often overshadowed by the poor programmability of FPGAs whose programming is conventionally a RTL design practice. Although recent advances in high-level synthesis (HLS) significantly improve the FPGA programmability, it still leaves programmers facing the challenge of identifying the optimal design configuration in a tremendous design space. This paper aims to address this challenge and pave the path from software programs towards high-quality FPGA accelerators. Specifically, we first propose the composable, parallel and pipeline (CPP) microarchitecture as a template of accelerator designs. Such a well-defined template is able to support efficient accelerator designs for a broad class of computation kernels, and more importantly, drastically reduce the design space. Also, we introduce an analytical model to capture the performance and resource trade-offs among different design configurations of the CPP microarchitecture, which lays the foundation for fast design space exploration. On top of the CPP microarchitecture and its analytical model, we develop the AutoAccel framework to make the entire accelerator generation automated. AutoAccel accepts a software program as an input and performs a series of code transformations based on the result of the analytical-model-based design space exploration to construct the desired CPP microarchitecture. Our experiments show that the AutoAccel-generated accelerators outperform their corresponding software implementations by an average of 72x for a broad class of computation kernels.
Sujit Manna, Peng Wei, Yingming Xie, Kam Tuen Law, Patrick Lee, Jagadeesh Moodera
Under certain conditions, a fermion in a superconductor can separate in space into two parts known as Majorana zero modes, which are immune to decoherence from local noise sources and are attractive building blocks for quantum computers. Promising experimental progress has been made to demonstrate Majorana zero modes in materials with strong spin-orbit coupling proximity coupled to superconductors. Here we report signatures of Majorana zero modes in a new material platform utilizing the surface states of gold. Using scanning tunneling microscope to probe EuS islands grown on top of gold nanowires, we observe two well separated zero bias tunneling conductance peaks aligned along the direction of the applied magnetic field, as expected for a pair of Majorana zero modes. This platform has the advantage of having a robust energy scale and the possibility of realizing complex designs using lithographic methods.
Peng Wei, Kun Guo, Ye Li, Jue Wang, Wei Feng, Shi Jin, Ning Ge, Ying-Chang Liang
Mobile edge computing (MEC) is considered a novel paradigm for computation-intensive and delay-sensitive tasks in fifth generation (5G) networks and beyond. However, its uncertainty, referred to as dynamic and randomness, from the mobile device, wireless channel, and edge network sides, results in high-dimensional, nonconvex, nonlinear, and NP-hard optimization problems. Thanks to the evolved reinforcement learning (RL), upon iteratively interacting with the dynamic and random environment, its trained agent can intelligently obtain the optimal policy in MEC. Furthermore, its evolved versions, such as deep RL (DRL), can achieve higher convergence speed efficiency and learning accuracy based on the parametric approximation for the large-scale state-action space. This paper provides a comprehensive research review on RL-enabled MEC and offers insight for development in this area. More importantly, associated with free mobility, dynamic channels, and distributed services, the MEC challenges that can be solved by different kinds of RL algorithms are identified, followed by how they can be solved by RL solutions in diverse mobile applications. Finally, the open challenges are discussed to provide helpful guidance for future research in RL training and learning MEC.
Cliff Chen, Protik Das, Ece Aytan, Weimin Zhou, Justin Horowitz, Biswarup Satpati, Alexander A. Balandin, Roger K. Lake, Peng Wei
The controlled tunability of superconductivity in low-dimensional materials may enable new quantum devices. Particularly in triplet or topological superconductors, tunneling devices such as Josephson junctions etc. can demonstrate exotic functionalities. The tunnel barrier, an insulating or normal material layer separating two superconductors, is a key component for the junctions. Thin layers of NbSe2 have been shown as a superconductor with strong spin orbit coupling, which can give rise to topological superconductivity if driven by a large magnetic exchange field. Here we demonstrate the superconductor-insulator transitions in epitaxially grown few-layer NbSe2 with wafer-scale uniformity on insulating substrates. We provide the electrical transport, Raman spectroscopy, cross-sectional transmission electron microscopy, and X-ray diffraction characterizations of the insulating phase. We show that the superconductor-insulator transition is driven by strain, which also causes characteristic energy shifts of the Raman modes. Our observation paves the way for high quality hetero-junction tunnel barriers to be seamlessly built into epitaxial NbSe2 itself, thereby enabling highly scalable tunneling devices for superconductor-based quantum electronics.
Wei Peng, Ehsan Adeli, Tomas Bosschieter, Sang Hyun Park, Qingyu Zhao, Kilian M. Pohl
As acquiring MRIs is expensive, neuroscience studies struggle to attain a sufficient number of them for properly training deep learning models. This challenge could be reduced by MRI synthesis, for which Generative Adversarial Networks (GANs) are popular. GANs, however, are commonly unstable and struggle with creating diverse and high-quality data. A more stable alternative is Diffusion Probabilistic Models (DPMs) with a fine-grained training strategy. To overcome their need for extensive computational resources, we propose a conditional DPM (cDPM) with a memory-efficient process that generates realistic-looking brain MRIs. To this end, we train a 2D cDPM to generate an MRI subvolume conditioned on another subset of slices from the same MRI. By generating slices using arbitrary combinations between condition and target slices, the model only requires limited computational resources to learn interdependencies between slices even if they are spatially far apart. After having learned these dependencies via an attention network, a new anatomy-consistent 3D brain MRI is generated by repeatedly applying the cDPM. Our experiments demonstrate that our method can generate high-quality 3D MRIs that share a similar distribution to real MRIs while still diversifying the training set. The code is available at https://github.com/xiaoiker/mask3DMRI_diffusion and also will be released as part of MONAI, at https://github.com/Project-MONAI/GenerativeModels.
Peng Wei, Ferhat Katmis, Cui-Zu Chang, Jagadeesh S. Moodera
We report a unique experimental approach to create topological superconductors by inducing superconductivity into epitaxial metallic thin film with strong spin-orbit coupling. Utilizing molecular beam epitaxy technique under ultra-high vacuum condition, we are able to achieve (111) oriented single phase of gold (Au) thin film grown on a well-oriented vanadium (V) s-wave superconductor film with clean interface. We obtained atomically smooth Au thin films with thicknesses even down to below a nanometer showing near-ideal surface quality. The as-grown V/Au bilayer heterostructure exhibits superconducting transition at around 4 K. Clear Josephson tunneling and Andreev reflection are observed in S-I-S tunnel junctions fabricated from the epitaxial bi-layers. The barrier thickness dependent tunneling and the associated subharmonic gap structures (SGS) confirmed the induced superconductivity in Au (111), paving the way for engineering thin film heterostructure based p-wave superconductors and nano devices for Majorana fermion.
Peng Wei
Numerous industrial processes can be defined using distributed parameter systems (DPSs). This study introduces a two-stage spatial construction approach for real-time modeling of DPSs in cases of limited sensors. Initially, a discrete space-completion approach is created to recuperate the spatiotemporal patterns of non-monitored locations under sparse sensing. The high-dimensional space construction method is employed to derive continuous spatial basis functions (SBFs). The identification and adjustment of the nonlinear temporal model are carried out via the long short-term memory (LSTM) neural network. Eventually, the amalgamation of the derived SBFs and temporal model results in a spatially continuous model. The use of a cubic B-spline surface is validated as an effective solution for optimizing space construction in the sense of least squares approximation. Experimental tests conducted on a pouch-type Li-ion battery demonstrate the efficacy of the proposed modeling technique under sparse sensing. This work highlights the promise of sparse sensors in real-time full-space modeling for large-scale battery energy storage systems.
Wei Peng, N. Lemée, J. -L. Dellis, V. V. Shvartsman, P. Borisov, W. Kleemann, Z. Trontelj, J. Holc, M. Kosec, R. Blinc, M. G. Karkut
We present electric and magnetic properties of 0.8Pb(Fe1/2Nb1/2)O3-0.2Pb(Mg1/2W1/2)O3 films epitaxially grown on (001) SrTiO3 substrates using pulsed laser deposition. A narrow deposition window around 710 oC and 0.2 mbar has been identified to achieve epitaxial single-phase thin films. A typical Vogel-Fulcher relaxor-like dielectric and magnetic susceptibility dispersion is observed, suggesting magnetoelectric relaxor behavior in these films similar to the bulk. We determine a magnetic cluster freezing temperature of 36 K, while observing weak ferromagnetism via magnetic hysteresis loops up to 300 K.
Wei Peng, Tomas Bosschieter, Jiahong Ouyang, Robert Paul, Ehsan Adeli, Qingyu Zhao, Kilian M. Pohl
Generative AI models hold great potential in creating synthetic brain MRIs that advance neuroimaging studies by, for example, enriching data diversity. However, the mainstay of AI research only focuses on optimizing the visual quality (such as signal-to-noise ratio) of the synthetic MRIs while lacking insights into their relevance to neuroscience. To gain these insights with respect to T1-weighted MRIs, we first propose a new generative model, BrainSynth, to synthesize metadata-conditioned (e.g., age- and sex-specific) MRIs that achieve state-of-the-art visual quality. We then extend our evaluation with a novel procedure to quantify anatomical plausibility, i.e., how well the synthetic MRIs capture macrostructural properties of brain regions, and how accurately they encode the effects of age and sex. Results indicate that more than half of the brain regions in our synthetic MRIs are anatomically accurate, i.e., with a small effect size between real and synthetic MRIs. Moreover, the anatomical plausibility varies across cortical regions according to their geometric complexity. As is, our synthetic MRIs can significantly improve the training of a Convolutional Neural Network to identify accelerated aging effects in an independent study. These results highlight the opportunities of using generative AI to aid neuroimaging research and point to areas for further improvement.
Wei Peng, Yue Hu, Luxi Xing, Yuqiang Xie, Yajing Sun, Yunpeng Li
Emotional support conversation aims at reducing the emotional distress of the help-seeker, which is a new and challenging task. It requires the system to explore the cause of help-seeker's emotional distress and understand their psychological intention to provide supportive responses. However, existing methods mainly focus on the sequential contextual information, ignoring the hierarchical relationships with the global cause and local psychological intention behind conversations, thus leads to a weak ability of emotional support. In this paper, we propose a Global-to-Local Hierarchical Graph Network to capture the multi-source information (global cause, local intentions and dialog history) and model hierarchical relationships between them, which consists of a multi-source encoder, a hierarchical graph reasoner, and a global-guide decoder. Furthermore, a novel training objective is designed to monitor semantic information of the global cause. Experimental results on the emotional support conversation dataset, ESConv, confirm that the proposed GLHG has achieved the state-of-the-art performance on the automatic and human evaluations. The code will be released in here \footnote{\small{~https://github.com/pengwei-iie/GLHG}}.
Wei Peng, Yue Hu, Luxi Xing, Yuqiang Xie, Xingsheng Zhang, Yajing Sun
Intention, emotion and action are important elements in human activities. Modeling the interaction process between individuals by analyzing the relationships between these elements is a challenging task. However, previous work mainly focused on modeling intention and emotion independently, and neglected of exploring the mutual relationships between intention and emotion. In this paper, we propose a RelAtion Interaction Network (RAIN), consisting of Intention Relation Module and Emotion Relation Module, to jointly model mutual relationships and explicitly integrate historical intention information. The experiments on the dataset show that our model can take full advantage of the intention, emotion and action between individuals and achieve a remarkable improvement over BERT-style baselines. Qualitative analysis verifies the importance of the mutual interaction between the intention and emotion.
Wei Peng, Yue Hu, Jing Yu, Luxi Xing, Yuqiang Xie, Zihao Zhu, Yajing Sun
Question answering systems usually use keyword searches to retrieve potential passages related to a question, and then extract the answer from passages with the machine reading comprehension methods. However, many questions tend to be unanswerable in the real world. In this case, it is significant and challenging how the model determines when no answer is supported by the passage and abstains from answering. Most of the existing systems design a simple classifier to determine answerability implicitly without explicitly modeling mutual interaction and relation between the question and passage, leading to the poor performance for determining the unanswerable questions. To tackle this problem, we propose a Multi-Step Co-Interactive Relation Network (MCR-Net) to explicitly model the mutual interaction and locate key clues from coarse to fine by introducing a co-interactive relation module. The co-interactive relation module contains a stack of interaction and fusion blocks to continuously integrate and fuse history-guided and current-query-guided clues in an explicit way. Experiments on the SQuAD 2.0 and DuReader datasets show that our model achieves a remarkable improvement, outperforming the BERT-style baselines in literature. Visualization analysis also verifies the importance of the mutual interaction between the question and passage.
Andrea Bonito, Peng Wei
We propose a finite element method for the numerical simulation of electroconvection of thin liquid crystals. The liquid is located in between two concentric circular electrodes which are either assumed to be of infinite height or slim. Each configuration results in a different nonlocal electro-magnetic model defined on a two dimensional bounded domain. The numerical method consists in approximating the surface charge density, the liquid velocity and pressure, and the electric potential in the two dimensional liquid region. Finite elements for the space discretization coupled with standard time stepping methods are put forward. Unlike for the infinite electrodes configuration, our numerical simulations indicate that slim electrodes are favorable for electroconvection to occur and are able to sustain the phenomena over long period of time. Furthermore, we provide a numerical study on the influence of the three main parameters of the system: the Rayleigh number, the Prandtl number and the electrodes aspect ratio.
Ting Han, Ximing Liu, Ryuichi Takanobu, Yixin Lian, Chongxuan Huang, Dazhen Wan, Wei Peng, Minlie Huang
Task-oriented dialogue systems have made unprecedented progress with multiple state-of-the-art (SOTA) models underpinned by a number of publicly available MultiWOZ datasets. Dialogue state annotations are error-prone, leading to sub-optimal performance. Various efforts have been put in rectifying the annotation errors presented in the original MultiWOZ dataset. In this paper, we introduce MultiWOZ 2.3, in which we differentiate incorrect annotations in dialogue acts from dialogue states, identifying a lack of co-reference when publishing the updated dataset. To ensure consistency between dialogue acts and dialogue states, we implement co-reference features and unify annotations of dialogue acts and dialogue states. We update the state of the art performance of natural language understanding and dialogue state tracking on MultiWOZ 2.3, where the results show significant improvements than on previous versions of MultiWOZ datasets (2.0-2.2).
Wei Peng, Li Feng, Guoying Zhao, Fang Liu
The inherent slow imaging speed of Magnetic Resonance Image (MRI) has spurred the development of various acceleration methods, typically through heuristically undersampling the MRI measurement domain known as k-space. Recently, deep neural networks have been applied to reconstruct undersampled k-space data and have shown improved reconstruction performance. While most of these methods focus on designing novel reconstruction networks or new training strategies for a given undersampling pattern, e.g., Cartesian undersampling or Non-Cartesian sampling, to date, there is limited research aiming to learn and optimize k-space sampling strategies using deep neural networks. This work proposes a novel optimization framework to learn k-space sampling trajectories by considering it as an Ordinary Differential Equation (ODE) problem that can be solved using neural ODE. In particular, the sampling of k-space data is framed as a dynamic system, in which neural ODE is formulated to approximate the system with additional constraints on MRI physics. In addition, we have also demonstrated that trajectory optimization and image reconstruction can be learned collaboratively for improved imaging efficiency and reconstruction performance. Experiments were conducted on different in-vivo datasets (e.g., brain and knee images) acquired with different sequences. Initial results have shown that our proposed method can generate better image quality in accelerated MRI than conventional undersampling schemes in Cartesian and Non-Cartesian acquisitions.
Wei Peng, Weien Zhou, Xiaoya Zhang, Wen Yao, Zheliang Liu
Learning solutions of partial differential equations (PDEs) with Physics-Informed Neural Networks (PINNs) is an attractive alternative approach to traditional solvers due to its flexibility and ease of incorporating observed data. Despite the success of PINNs in accurately solving a wide variety of PDEs, the method still requires improvements in terms of computational efficiency. One possible improvement idea is to optimize the generation of training point sets. Residual-based adaptive sampling and quasi-uniform sampling approaches have been each applied to improve the training effects of PINNs, respectively. To benefit from both methods, we propose the Residual-based Adaptive Node Generation (RANG) approach for efficient training of PINNs, which is based on a variable density nodal distribution method for RBF-FD. The method is also enhanced by a memory mechanism to further improve training stability. We conduct experiments on three linear PDEs and three nonlinear PDEs with various node generation methods, through which the accuracy and efficiency of the proposed method compared to the predominant uniform sampling approach is verified numerically.
Wei Peng, Yuhong Dai, Hui Zhang, Lizhi Cheng
Training generative adversarial networks (GANs) often suffers from cyclic behaviors of iterates. Based on a simple intuition that the direction of centripetal acceleration of an object moving in uniform circular motion is toward the center of the circle, we present the Simultaneous Centripetal Acceleration (SCA) method and the Alternating Centripetal Acceleration (ACA) method to alleviate the cyclic behaviors. Under suitable conditions, gradient descent methods with either SCA or ACA are shown to be linearly convergent for bilinear games. Numerical experiments are conducted by applying ACA to existing gradient-based algorithms in a GAN setup scenario, which demonstrate the superiority of ACA.
Peng Wei, Yue Xiao, Wei Xiang
A novel basis signal optimization method is proposed for reducing the interference in the N-continuous orthogonal frequency division multiplexing (NC-OFDM) system. Compared to conventional NC-OFDM, the proposed scheme is capable of improving the transmission performance while maintaining an identical sidelobe suppression performance imposed by the linear combination of two groups of basis signals. Our performance results demonstrate that with a low complexity overhead, the proposed scheme is capable of striking a better trade-off among the bit error rate (BER), complexity, and the sidelobe suppression performance compared to its conventional counterpart.
Wei Peng, Xiaopeng Hong, Guoying Zhao
Deep neural networks have achieved great success for video analysis and understanding. However, designing a high-performance neural architecture requires substantial efforts and expertise. In this paper, we make the first attempt to let algorithm automatically design neural networks for video action recognition tasks. Specifically, a spatio-temporal network is developed in a differentiable space modeled by a directed acyclic graph, thus a gradient-based strategy can be performed to search an optimal architecture. Nonetheless, it is computationally expensive, since the computational burden to evaluate each architecture candidate is still heavy. To alleviate this issue, we, for the video input, introduce a temporal segment approach to reduce the computational cost without losing global video information. For the architecture, we explore in an efficient search space by introducing pseudo 3D operators. Experiments show that, our architecture outperforms popular neural architectures, under the training from scratch protocol, on the challenging UCF101 dataset, surprisingly, with only around one percentage of parameters of its manual-design counterparts.