arXiv subjects

Wei Lu

Publications and source records attributed to Wei Lu.

At least 289 records · Page 16Linked to original sources

arXiv2022

Ultrafast photothermoelectric effect in Dirac semimetallic Cd3As2 revealed by terahertz emission

The thermoelectric effects of topological semimetals have attracted tremendous research interest because many topological semimetals are excellent thermoelectric materials and thermoelectricity serves as one of their most important potential applications. In this work, we reveal the transient photothermoelectric response of Dirac semimetallic Cd3As2, namely the photo-Seebeck effect and photo-Nernst effect, by studying the terahertz (THz) emission from the transient photocurrent induced by these effects. Our excitation polarization and power dependence confirm that the observed THz emission is due to photothermoelectric effect instead of other nonlinear optical effect. Furthermore, when a weak magnetic field (~0.4 T) is applied, the response clearly indicates an order of magnitude enhancement on transient photothermoelectric current generation compared to the photo-Seebeck effect. Such enhancement supports an ambipolar transport nature of the photo-Nernst current generation in Cd3As2. These results highlight the enhancement of thermoelectric performance can be achieved in topological Dirac semimetals based on the Nernst effect, and our transient studies pave the way for thermoelectric devices applicable for high field circumstance when nonequilibrium state matters. The large THz emission due to highly efficient photothermoelectric conversion is comparable to conventional semiconductors through optical rectification and photo-Dember effect.

cond-mat.mtrl-sci↗

arXiv2022

Compact polarized X-ray source based on all-optical inverse Compton scattering

Polarized X-ray source is an important probe for many fields such as fluorescence imaging, magnetic microscopy, and nuclear physics research. All-optical inverse Compton scattering source (AOCS) based on laser wakefield accelerator (LWFA) has drawn great attention in recent years due to its compact scale and high performance, especially its potential to generate polarized X-rays. Here, polarization-tunable X-rays are generated by a plasma-mirror-based AOCS scheme. The linearly and circularly polarized AOCS pulses are achieved with the mean photon energy of 60($\pm$5)/64($\pm$3) keV and the single-shot photon yield of $\sim$1.1/1.3$\times10^7$. A Compton polarimeter is designed to diagnose the photon polarization states, demonstrating AOCS's polarization-tunable property, and indicating the average polarization degree of the linearly polarized AOCS is 75($\pm$3)%.

physics.app-ph↗

arXiv2022

Quark-Gluon Plasma and Nucleons a la Laughlin

Inspired by Laughlin's theory of the fractional quantum Hall effect, we explore the topological nature of the quark-gluon plasma and the nucleons in the context of the Clifford algebra. In our model, each quark is transformed into a composite particle via the simultaneous attachment of a spin monopole and an isospin monopole. This is induced by a novel kind of meson endowed with both spin and isospin degrees of freedom. The interactions in the strongly coupled quark-gluon system are governed by the topological winding number of the monopoles, which is an odd integer to ensure that the overall wave function is antisymmetric. The states of the quark-gluon plasma and the nucleons are thus uniquely determined by the combination of the monopole winding number m and the total quark number N. The radius squared of the quark-gluon plasma droplet is expected to be proportional to mN. We anticipate the observation of such proportionality in the heavy ion collision experiments.

hep-ph↗

arXiv2022

Self-Directed Online Machine Learning for Topology Optimization

Topology optimization by optimally distributing materials in a given domain requires non-gradient optimizers to solve highly complicated problems. However, with hundreds of design variables or more involved, solving such problems would require millions of Finite Element Method (FEM) calculations whose computational cost is huge and impractical. Here we report Self-directed Online Learning Optimization (SOLO) which integrates Deep Neural Network (DNN) with FEM calculations. A DNN learns and substitutes the objective as a function of design variables. A small number of training data is generated dynamically based on the DNN's prediction of the optimum. The DNN adapts to the new training data and gives better prediction in the region of interest until convergence. The optimum predicted by the DNN is proved to converge to the true global optimum through iterations. Our algorithm was tested by four types of problems including compliance minimization, fluid-structure optimization, heat transfer enhancement and truss optimization. It reduced the computational time by 2 ~ 5 orders of magnitude compared with directly using heuristic methods, and outperformed all state-of-the-art algorithms tested in our experiments. This approach enables solving large multi-dimensional optimization problems.

cs.CE↗

arXiv2021

Sample Average Approximation for Stochastic Optimization with Dependent Data: Performance Guarantees and Tractability

Sample average approximation (SAA), a popular method for tractably solving stochastic optimization problems, enjoys strong asymptotic performance guarantees in settings with independent training samples. However, these guarantees are not known to hold generally with dependent samples, such as in online learning with time series data or distributed computing with Markovian training samples. In this paper, we show that SAA remains tractable when the distribution of unknown parameters is only observable through dependent instances and still enjoys asymptotic consistency and finite sample guarantees. Specifically, we provide a rigorous probability error analysis to derive $1 - β$ confidence bounds for the out-of-sample performance of SAA estimators and show that these estimators are asymptotically consistent. We then, using monotone operator theory, study the performance of a class of stochastic first-order algorithms trained on a dependent source of data. We show that approximation error for these algorithms is bounded and concentrates around zero, and establish deviation bounds for iterates when the underlying stochastic process is $ϕ$-mixing. The algorithms presented can be used to handle numerically inconvenient loss functions such as the sum of a smooth and non-smooth function or of non-smooth functions with constraints. To illustrate the usefulness of our results, we present several stochastic versions of popular algorithms such as stochastic proximal gradient descent (S-PGD), stochastic relaxed Peaceman--Rachford splitting algorithms (S-rPRS), and numerical experiment.

math.OC↗

arXiv2021

High efficiency uniform positron beam loading in a hollow channel plasma wakefield accelerator

We propose a novel positron beam loading regime in a hollow plasma channel that can efficiently accelerate $e^+$ beam with high gradient and narrow energy spread. In this regime, the $e^+$ beam coincides with the drive $e^-$ beam in time and space and their net current distribution determines the plasma wakefields. By precisely shaping the beam current profile and loading phase according to explicit expressions, three-dimensional Particle-in-Cell (PIC) simulations show that the acceleration for $e^+$ beam of $\sim$nC charge with $\sim$GV/m gradient, $\lesssim$0.5% induced energy spread and $\sim$50% energy transfer efficiency can be achieved simultaneously. Besides, only tailoring the current profile of the more tunable $e^-$ beam instead of the $e^+$ beam is enough to obtain such favorable results. A theoretical analysis considering both linear and nonlinear plasma responses in hollow plasma channels is proposed to quantify the beam loading effects. This theory agrees very well with the simulation results and verifies the robustness of this beam loading regime over a wide range of parameters.

physics.plasm-ph↗

arXiv2021

Coo: Rethink Data Anomalies In Databases

Transaction processing technology has three important contents: data anomalies, isolation levels, and concurrent control algorithms. Concurrent control algorithms are used to eliminate some or all data anomalies at different isolation levels to ensure data consistency. Isolation levels in the current ANSI standard are defined by disallowing certain kinds of data anomalies. Yet, the definitions of data anomalies in the ANSI standard are controversial. On one hand, the definitions lack a mathematical formalization and cause ambiguous interpretations. On the other hand, the definitions are made in a case-by-case manner and lead to a situation that even a senior DBA could not have infallible knowledge of data anomalies, due to a lack of a full understanding of its nature. While revised definitions in existing literature propose various mathematical formalizations to correct the former argument, how to address the latter argument still remains an open problem. In this paper, we present a general framework called Coo with the capability to systematically define data anomalies. Under this framework, we show that existing reported data anomalies are only a small portion. While we theoretically prove that Coo is complete to mathematically formalize data anomalies, we employ a novel method to classify infinite data anomalies. In addition, we use this framework to define new isolation levels and quantitatively describe the concurrency and rollback rate of mainstream concurrency control algorithms. These works show that the C and I of ACID can be quantitatively analyzed based on all data anomalies.

cs.DB↗

arXiv2021

Speaker-Oriented Latent Structures for Dialogue-Based Relation Extraction

Dialogue-based relation extraction (DiaRE) aims to detect the structural information from unstructured utterances in dialogues. Existing relation extraction models may be unsatisfactory under such a conversational setting, due to the entangled logic and information sparsity issues in utterances involving multiple speakers. To this end, we introduce SOLS, a novel model which can explicitly induce speaker-oriented latent structures for better DiaRE. Specifically, we learn latent structures to capture the relationships among tokens beyond the utterance boundaries, alleviating the entangled logic issue. During the learning process, our speaker-specific regularization method progressively highlights speaker-related key clues and erases the irrelevant ones, alleviating the information sparsity issue. Experiments on three public datasets demonstrate the effectiveness of our proposed approach.

cs.CL↗

arXiv2021

Exploring Task Difficulty for Few-Shot Relation Extraction

Few-shot relation extraction (FSRE) focuses on recognizing novel relations by learning with merely a handful of annotated instances. Meta-learning has been widely adopted for such a task, which trains on randomly generated few-shot tasks to learn generic data representations. Despite impressive results achieved, existing models still perform suboptimally when handling hard FSRE tasks, where the relations are fine-grained and similar to each other. We argue this is largely because existing models do not distinguish hard tasks from easy ones in the learning process. In this paper, we introduce a novel approach based on contrastive learning that learns better representations by exploiting relation label information. We further design a method that allows the model to adaptively learn how to focus on hard tasks. Experiments on two standard datasets demonstrate the effectiveness of our method.

cs.CL↗

arXiv2021

A Minimal Physics-Based Model on the Electrochemical Impedance Spectroscopy of Solid-State Electrolyte

Solid state batteries have emerged as a potential next-generation energy storage device due to safety and energy density advantages. Development of electrolyte is one of the most important topics in solid state batteries. Electrochemical Impedance Spectroscopy (EIS) is a popular measurement technique to obtain the conductivity and diagnose the electrolyte. Current interpretation mainly uses the semicircle part of the curves and discards other information revealed by EIS such as the slope of the curve at low frequency. What is worse, some features on the curve are not fully interpreted. To better understand the transport mechanism and interpret EIS curves, we introduce a continuous model to quantify the ion transport and current flow in the electrolyte. The produced EIS curves from the model are compared with experiment data to show good agreement.

cond-mat.mtrl-sci↗

arXiv2021

Profiling Neural Blocks and Design Spaces for Mobile Neural Architecture Search

Neural architecture search automates neural network design and has achieved state-of-the-art results in many deep learning applications. While recent literature has focused on designing networks to maximize accuracy, little work has been conducted to understand the compatibility of architecture design spaces to varying hardware. In this paper, we analyze the neural blocks used to build Once-for-All (MobileNetV3), ProxylessNAS and ResNet families, in order to understand their predictive power and inference latency on various devices, including Huawei Kirin 9000 NPU, RTX 2080 Ti, AMD Threadripper 2990WX, and Samsung Note10. We introduce a methodology to quantify the friendliness of neural blocks to hardware and the impact of their placement in a macro network on overall network performance via only end-to-end measurements. Based on extensive profiling results, we derive design insights and apply them to hardware-specific search space reduction. We show that searching in the reduced search space generates better accuracy-latency Pareto frontiers than searching in the original search spaces, customizing architecture search according to the hardware. Moreover, insights derived from measurements lead to notably higher ImageNet top-1 scores on all search spaces investigated.

cs.LG↗

arXiv2021

L$^{2}$NAS: Learning to Optimize Neural Architectures via Continuous-Action Reinforcement Learning

Neural architecture search (NAS) has achieved remarkable results in deep neural network design. Differentiable architecture search converts the search over discrete architectures into a hyperparameter optimization problem which can be solved by gradient descent. However, questions have been raised regarding the effectiveness and generalizability of gradient methods for solving non-convex architecture hyperparameter optimization problems. In this paper, we propose L$^{2}$NAS, which learns to intelligently optimize and update architecture hyperparameters via an actor neural network based on the distribution of high-performing architectures in the search history. We introduce a quantile-driven training procedure which efficiently trains L$^{2}$NAS in an actor-critic framework via continuous-action reinforcement learning. Experiments show that L$^{2}$NAS achieves state-of-the-art results on NAS-Bench-201 benchmark as well as DARTS search space and Once-for-All MobileNetV3 search space. We also show that search policies generated by L$^{2}$NAS are generalizable and transferable across different training datasets with minimal fine-tuning.

cs.LG↗

arXiv2021

To be Closer: Learning to Link up Aspects with Opinions

Dependency parse trees are helpful for discovering the opinion words in aspect-based sentiment analysis (ABSA). However, the trees obtained from off-the-shelf dependency parsers are static, and could be sub-optimal in ABSA. This is because the syntactic trees are not designed for capturing the interactions between opinion words and aspect words. In this work, we aim to shorten the distance between aspects and corresponding opinion words by learning an aspect-centric tree structure. The aspect and opinion words are expected to be closer along such tree structure compared to the standard dependency parse tree. The learning process allows the tree structure to adaptively correlate the aspect and opinion words, enabling us to better identify the polarity in the ABSA task. We conduct experiments on five aspect-based sentiment datasets, and the proposed model significantly outperforms recent strong baselines. Furthermore, our thorough analysis demonstrates the average distance between aspect and opinion words are shortened by at least 19% on the standard SemEval Restaurant14 dataset.

cs.CL↗

arXiv2021

A Role-Selected Sharing Network for Joint Machine-Human Chatting Handoff and Service Satisfaction Analysis

Chatbot is increasingly thriving in different domains, however, because of unexpected discourse complexity and training data sparseness, its potential distrust hatches vital apprehension. Recently, Machine-Human Chatting Handoff (MHCH), predicting chatbot failure and enabling human-algorithm collaboration to enhance chatbot quality, has attracted increasing attention from industry and academia. In this study, we propose a novel model, Role-Selected Sharing Network (RSSN), which integrates both dialogue satisfaction estimation and handoff prediction in one multi-task learning framework. Unlike prior efforts in dialog mining, by utilizing local user satisfaction as a bridge, global satisfaction detector and handoff predictor can effectively exchange critical information. Specifically, we decouple the relation and interaction between the two tasks by the role information after the shared encoder. Extensive experiments on two public datasets demonstrate the effectiveness of our model.

cs.CL↗

arXiv2021

Uncovering Main Causalities for Long-tailed Information Extraction

Information Extraction (IE) aims to extract structural information from unstructured texts. In practice, long-tailed distributions caused by the selection bias of a dataset, may lead to incorrect correlations, also known as spurious correlations, between entities and labels in the conventional likelihood models. This motivates us to propose counterfactual IE (CFIE), a novel framework that aims to uncover the main causalities behind data in the view of causal inference. Specifically, 1) we first introduce a unified structural causal model (SCM) for various IE tasks, describing the relationships among variables; 2) with our SCM, we then generate counterfactuals based on an explicit language structure to better calculate the direct causal effect during the inference stage; 3) we further propose a novel debiasing approach to yield more robust predictions. Experiments on three IE tasks across five public datasets show the effectiveness of our CFIE model in mitigating the spurious correlation issues.

cs.CL↗

arXiv2021

A Facile Process to Fabricate Phosphorus/Carbon Xerogel Composite as Anode for Sodium Ion Batteries

Sodium ion batteries become popular due to their low cost. Among possible anode materials of sodium ion batteries, phosphorus has great potential owing to its high theoretical capacity. Previous research that yielded high capacity and long duration of phosphorus anode used expensive materials such as black phosphorus (BP) and phosphorene. To take advantage of the low cost of sodium ion batteries, we report a simple and low-cost method to fabricate anode: condensing red phosphorus on carbon xerogel. Even with large particle size (~ 50 $μ$m) and high mass loading (2 mg cm$^{-2}$), the composite cycled at 100 mA g$^{-1}$ yielded a capacity of 357 mA g$^{-1}$ or 2498 mAh g$^{-1}_P$ based on phosphorus after subtracting the contribution of carbon. The average coulombic efficiency is as high as 99.4%. When cycled at 200 mA g$^{-1}$, it yielded a capacity of 242 mAh g$^{-1}$ or 1723 mAh g$^{-1}_P$, with average degradation rate only 0.06% in 80 cycles. Our research provided an innovative approach to synthesize anodes for sodium ion batteries at extremely low cost, with performance exceeding or comparable to state-of-the-art materials, which will promote their commercialization.

physics.app-ph↗

arXiv2021

Interventional Video Grounding with Dual Contrastive Learning

Video grounding aims to localize a moment from an untrimmed video for a given textual query. Existing approaches focus more on the alignment of visual and language stimuli with various likelihood-based matching or regression strategies, i.e., P(Y|X). Consequently, these models may suffer from spurious correlations between the language and video features due to the selection bias of the dataset. 1) To uncover the causality behind the model and data, we first propose a novel paradigm from the perspective of the causal inference, i.e., interventional video grounding (IVG) that leverages backdoor adjustment to deconfound the selection bias based on structured causal model (SCM) and do-calculus P(Y|do(X)). Then, we present a simple yet effective method to approximate the unobserved confounder as it cannot be directly sampled from the dataset. 2) Meanwhile, we introduce a dual contrastive learning approach (DCL) to better align the text and video by maximizing the mutual information (MI) between query and video clips, and the MI between start/end frames of a target moment and the others within a video to learn more informative visual representations. Experiments on three standard benchmarks show the effectiveness of our approaches. Our code is available on GitHub: https://github.com/nanguoshun/IVG.

cs.CV↗

arXiv2021

Mixed Cross Entropy Loss for Neural Machine Translation

In neural machine translation, cross entropy (CE) is the standard loss function in two training methods of auto-regressive models, i.e., teacher forcing and scheduled sampling. In this paper, we propose mixed cross entropy loss (mixed CE) as a substitute for CE in both training approaches. In teacher forcing, the model trained with CE regards the translation problem as a one-to-one mapping process, while in mixed CE this process can be relaxed to one-to-many. In scheduled sampling, we show that mixed CE has the potential to encourage the training and testing behaviours to be similar to each other, more effectively mitigating the exposure bias problem. We demonstrate the superiority of mixed CE over CE on several machine translation datasets, WMT'16 Ro-En, WMT'16 Ru-En, and WMT'14 En-De in both teacher forcing and scheduled sampling setups. Furthermore, in WMT'14 En-De, we also find mixed CE consistently outperforms CE on a multi-reference set as well as a challenging paraphrased reference set. We also found the model trained with mixed CE is able to provide a better probability distribution defined over the translation output space. Our code is available at https://github.com/haorannlp/mix.

cs.CL↗