arXiv subjects

Yan Gao

Publications and source records attributed to Yan Gao.

At least 199 records · Page 11Linked to original sources

arXiv2021

Characterizing Shadow Price via Lagrangian Multiplier for Nonsmooth Problem

In this paper, a relation between shadow price and the Lagrangian multiplier for nonsmooth problem is explored. It is shown that the Lagrangian Multiplier is the upper bound of shadow price for convex optimization and a class of Lipschtzian optimizations. This work can be used in shadow pricing for nonsmooth situation. The several nonsmooth functions involved in this class of Lipschtzian optimizations is listed. Finally, an application to electricity pricing is discussed.

econ.TH↗

arXiv2021

Intrinsic ferromagnetic and antiferromagnetic axion insulators in van der Waals materials Mn\emph{X}$_{2}$\emph{B}$_{2}$\emph{T}$_{6}$ family

The MnBi$_{2}$Te$_{4}$ family has attracted significant attention due to its rich topological states such as the quantum anomalous Hall (QAH) insulator state, the axion insulator state, and the magnetic Weyl semimetal state. Nevertheless, the intrinsic antiferromagnetic (AFM) interlayer coupling in MnBi$_{2}$Te$_{4}$ partly hinders the realization of "high-temperature" QAH effect. Here, by using first-principles electronic structure calculations, we design a new class of materials Mn\emph{X}$_{2}$\emph{B}$_{2}$\emph{T}$_{6}$ (\emph{X}=Ge, Sn, or Pb; \emph{B}=Sb or Bi; \emph{T}=Se or Te) based on the \emph{X}$_{2}$\emph{B}$_{2}$\emph{T}$_{5}$ structures rather than the Bi$_{2}$Te$_{3}$ family. We find that each septuple-layer Mn\emph{B}$_{2}$\emph{T}$_{4}$ is sandwiched by two [\emph{X}\emph{T}] layers, which may turn the AFM interlayer coupling into a ferromagnetic (FM) coupling. The calculations specifically demonstrate that \emph{MnGe}$_{2}$\emph{Sb}$_{2}$\emph{Te}$_{6}$, \emph{MnGe}$_{2}$\emph{Bi}$_{2}$\emph{Te}$_{6}$, and \emph{MnPb}$_{2}$\emph{Bi}$_{2}$\emph{Te}$_{6}$ are FM axion insulators, while MnGe$_{2}$Sb$_{2}$Se$_{6}$, MnGe$_{2}$Bi$_{2}$Se$_{6}$, MnSn$_{2}$Sb$_{2}$Te$_{6}$, and MnSn$_{2}$Bi$_{2}$Te$_{6}$ are A-type AFM axion insulators. These seven materials all have an out-of-plane easy axis of magnetization. The Mn\emph{X}$_{2}$\emph{B}$_{2}$\emph{T}$_{6}$ family thus offers a promising platform beyond the MnBi$_{2}$Te$_{4}$ family for the realization of quantized magnetoelectric effect and "high-temperature" QAH effect in future experiments.

cond-mat.mtrl-sci↗

arXiv2021

Occluded Video Instance Segmentation: Dataset and ICCV 2021 Challenge

Although deep learning methods have achieved advanced video object recognition performance in recent years, perceiving heavily occluded objects in a video is still a very challenging task. To promote the development of occlusion understanding, we collect a large-scale dataset called OVIS for video instance segmentation in the occluded scenario. OVIS consists of 296k high-quality instance masks and 901 occluded scenes. While our human vision systems can perceive those occluded objects by contextual reasoning and association, our experiments suggest that current video understanding systems cannot. On the OVIS dataset, all baseline methods encounter a significant performance degradation of about 80% in the heavily occluded object group, which demonstrates that there is still a long way to go in understanding obscured objects and videos in a complex real-world scenario. To facilitate the research on new paradigms for video understanding systems, we launched a challenge based on the OVIS dataset. The submitted top-performing algorithms have achieved much higher performance than our baselines. In this paper, we will introduce the OVIS dataset and further dissect it by analyzing the results of baselines and submitted methods. The OVIS dataset and challenge information can be found at http://songbai.site/ovis .

cs.CV↗

arXiv2021

Generation of entanglement between a highly wave-packet-tunable photon and a spin-wave memory in cold atoms

Controls of waveforms (pulse durations) of single photons are important tasks for effectively interconnecting disparate atomic memories in hybrid quantum networks. So far, the waveform control of single photon that is entangled with an atomic memory remains unexplored. Here, we demonstrated control of waveform length of the photon that is entangled with an atomic spin-wave memory by varying light-atom interaction time in cold atoms. The Bell parameter S as a function of the duration of photon pulse is measured, which shows that violations of Bell equality can be achieved for the photon pulse in the duration range from 40 ns to 50 us, where, S=2.64+/-0.02 and S=2.26+/-0.05 for the 40-ns and 50-μs durations, respectively. The measured results show that S parameter decreases with the increase in the pulse duration. We confirm that the increase in photon noise probability per pulse with the pulse-duration is responsible for the S decrease.

quant-ph↗

arXiv2021

Practical Distributed Control for VTOL UAVs to Pass a Virtual Tube

Unmanned Aerial Vehicles (UAVs) are now becoming increasingly accessible to amateur and commercial users alike. An air traffic management (ATM) system is needed to help ensure that this newest entrant into the skies does not collide with others. In an ATM, airspace can be composed of airways, intersections and nodes. In this paper, for simplicity, distributed coordinating the motions of Vertical TakeOff and Landing (VTOL) UAVs to pass an airway is focused. This is formulated as a virtual tube passing problem, which includes passing a virtual tube, inter-agent collision avoidance and keeping within the virtual tube. Lyapunov-like functions are designed elaborately, and formal analysis based on invariant set theorem is made to show that all UAVs can pass the virtual tube without getting trapped, avoid collision and keep within the virtual tube. What is more, by the proposed distributed control, a VTOL UAV can keep away from another VTOL UAV or return back to the virtual tube as soon as possible, once it enters into the safety area of another or has a collision with the virtual tube during it is passing the virtual tube. Simulations and experiments are carried out to show the effectiveness of the proposed method and the comparison with other methods.

cs.RO↗

arXiv2021

A Catalogue of Potential Post Common Envelope Binaries

We present a catalogue containing 839 candidate post common envelope systems. Common envelope evolution is very important in stellar astrophysics, particularly in the context of very compact and short-period binaries, including cataclysmic variables, as progenitors of e.g. supernovae type Ia or mergers of black holes and/or neutron stars. At the same time it is a barely understood process in binary evolution. Due to limitations, since partially remedied, on direct simulation, early investigations were mainly focused on providing analytic prescriptions of the outcome of common envelope evolution. In recent years, detailed hydrodynamical calculations have produced deeper insight into the previously elusive process of envelope ejection. However, a direct link between observations and theory of this relatively short-lived phase in binary evolution has not been forthcoming. Therefore, the main insight to be gained from observations has to be derived from the current state of systems likely to have gone through a common envelope. Here we present an extensive catalogue of such observations as found in the literature. The aim of this paper is to provide a reliable set of data, obtained from observations, to be used in the theoretical modelling of common envelope evolution. In this catalogue, the former common envelope donor star is commonly observed as a white dwarf star or as a hot sub-dwarf star. This catalogue includes period and mass estimates, wherever obtainable. Some binaries are border line cases to allow an investigation of the transition between a common envelope formation and other mass-transfer processes.

astro-ph.SR↗

arXiv2021

End-to-End Speech Recognition from Federated Acoustic Models

Training Automatic Speech Recognition (ASR) models under federated learning (FL) settings has attracted a lot of attention recently. However, the FL scenarios often presented in the literature are artificial and fail to capture the complexity of real FL systems. In this paper, we construct a challenging and realistic ASR federated experimental setup consisting of clients with heterogeneous data distributions using the French and Italian sets of the CommonVoice dataset, a large heterogeneous dataset containing thousands of different speakers, acoustic environments and noises. We present the first empirical study on attention-based sequence-to-sequence End-to-End (E2E) ASR model with three aggregation weighting strategies -- standard FedAvg, loss-based aggregation and a novel word error rate (WER)-based aggregation, compared in two realistic FL scenarios: cross-silo with 10 clients and cross-device with 2K and 4K clients. Our analysis on E2E ASR from heterogeneous and realistic federated acoustic models provides the foundations for future research and development of realistic FL-based ASR applications.

cs.SD↗

arXiv2021

Distilling Knowledge from Ensembles of Acoustic Models for Joint CTC-Attention End-to-End Speech Recognition

Knowledge distillation has been widely used to compress existing deep learning models while preserving the performance on a wide range of applications. In the specific context of Automatic Speech Recognition (ASR), distillation from ensembles of acoustic models has recently shown promising results in increasing recognition performance. In this paper, we propose an extension of multi-teacher distillation methods to joint CTC-attention end-to-end ASR systems. We also introduce three novel distillation strategies. The core intuition behind them is to integrate the error rate metric to the teacher selection rather than solely focusing on the observed losses. In this way, we directly distill and optimize the student toward the relevant metric for speech recognition. We evaluate these strategies under a selection of training procedures on different datasets (TIMIT, Librispeech, Common Voice) and various languages (English, French, Italian). In particular, state-of-the-art error rates are reported on the Common Voice French, Italian and TIMIT datasets.

cs.LG↗

arXiv2021

Theoretical design of all-carbon networks with intrinsic magnetism

To induce intrinsic magnetism in the nominally nonmagnetic carbon materials containing only $s$ and $p$ electrons is an intriguing yet challenging task. Here, based on first-principles electronic structure calculations, we propose a universal approach inspired by Ovchinnikov's rule to guide us the design of a series of imaginative magnetic all-carbon structures. The idea is to combine the differently stacked graphene layers via the acetylenic linkages (-C$\equiv$C-) to obtain a class of two-dimensional (2D) and three-dimensional (3D) carbon networks. With first-principles electronic structure calculations, we confirm the effectiveness of this approach via concrete examples of double-layer ALBG-C14, triple-layer ALTG-C22, and bulk IALG-C30. We show that these materials are antiferromagnetic (AFM) semiconductors with intralayer Néel and interlayer AFM couplings. According to the above idea, our work not only provides a promising design scheme for magnetic all-carbon materials, but also can apply to other $π$-bonding network systems.

cond-mat.mtrl-sci↗

arXiv2021

$R$RuB$_{2}$ ($R$=Y, Lu), topological superconductor candidates with hourglass-type Dirac ring

Topological properties and topological superconductivity in real materials have attracted intensive experimental and theoretical attention recently. Based on symmetry analysis and first-principles electronic structure calculations, we predict that $R$RuB$_{2}$ ($R$=Y, Lu) are not only topological superconductor (TSC) candidates, but also own the hybrid hourglass-type Dirac ring which is protected by the nonsymmorphic space group symmetry. Due to the band inversion around the time-reversal invariant $Γ$ point in the Brillouin zone,$R$RuB$_{2}$ also have Dirac topological surface states (TSSs). More importantly, their TSSs on the (010) surface are within the band gap of bulk and cross the Fermi level, which form single Fermi surfaces. Considering the fact that both YRuB$_{2}$ and LuRuB$_{2}$ are superconductors with respective superconducting transition temperatures ($T_c$) of 7.6 K and 10.2 K, the superconducting bulks will likely induce superconductivity in the TSSs via the proximity effect. The ternary borides $R$RuB$_{2}$ may thus provide a very promising platform for studying the properties of topological superconductivity and hourglass fermions in the future experiments.

cond-mat.supr-con↗

arXiv2021

SpeechBrain: A General-Purpose Speech Toolkit

SpeechBrain is an open-source and all-in-one speech toolkit. It is designed to facilitate the research and development of neural speech processing technologies by being simple, flexible, user-friendly, and well-documented. This paper describes the core architecture designed to support several tasks of common interest, allowing users to naturally conceive, compare and share novel speech processing pipelines. SpeechBrain achieves competitive or state-of-the-art performance in a wide range of speech benchmarks. It also provides training recipes, pretrained models, and inference scripts for popular speech datasets, as well as tutorials which allow anyone with basic Python proficiency to familiarize themselves with speech technologies.

eess.AS↗

arXiv2021

On-device Federated Learning with Flower

Federated Learning (FL) allows edge devices to collaboratively learn a shared prediction model while keeping their training data on the device, thereby decoupling the ability to do machine learning from the need to store data in the cloud. Despite the algorithmic advancements in FL, the support for on-device training of FL algorithms on edge devices remains poor. In this paper, we present an exploration of on-device FL on various smartphones and embedded devices using the Flower framework. We also evaluate the system costs of on-device FL and discuss how this quantification could be used to design more efficient FL algorithms.

cs.LG↗

arXiv2021

Hyperspectral Image Denoising Based On Multi-Stream Denoising Network

Hyperspectral images (HSIs) have been widely applied in many fields, such as military, agriculture, and environment monitoring. Nevertheless, HSIs commonly suffer from various types of noise during acquisition. Therefore, denoising is critical for HSI analysis and applications. In this paper, we propose a novel blind denoising method for HSIs based on Multi-Stream Denoising Network (MSDNet). Our network consists of the noise estimation subnetwork and denoising subnetwork. In the noise estimation subnetwork, a multiscale fusion module is designed to capture the noise from different scales. Then, the denoising subnetwork is utilized to obtain the final denoising image. The proposed MSDNet can obtain robust noise level estimation, which is capable of improving the performance of HSI denoising. Extensive experiments on HSI dataset demonstrate that the proposed method outperforms four closely related methods.

eess.IV↗

arXiv2020

Topological properties of non-centrosymmetric superconductors $T$Ir$_{2}$B$_{2}$ ($T$=Nb, Ta)

A recent experiment reported two new non-centrosymmetric superconductors NbIr$_{2}$B$_{2}$ and TaIr$_{2}$B$_{2}$ with respective superconducting transition temperatures of 7.2 K and 5.2 K and further suggested their superconductivity to be unconventional [K. Górnicka \textit{et al}., Adv. Funct. Mater. 2007960 (2020)]. Here, based on first-principles calculations and symmetry analysis, we propose that $T$Ir$_{2}$B$_{2}$ ($T$=Nb, Ta) are topological Weyl metals in the normal state. In the absence of spin-orbit coupling (SOC), we find that NbIr$_{2}$B$_{2}$ has 12 Weyl points, and TaIr$_{2}$B$_{2}$ has 4 Weyl points, i.e. the minimum number under time-reversal symmetry; meanwhile, both of them have a nodal net composed of three nodal lines. In the presence of SOC, a nodal loop on the mirror plane evolves into two hourglass Weyl rings, along with the Weyl points, which are dictated by the nonsymmorphic glide mirror symmetry. Besides the rings, NbIr$_{2}$B$_{2}$ and TaIr$_{2}$B$_{2}$ have 16 and 20 pairs of Weyl points, respectively. The surface Fermi arcs are explicitly demonstrated. On the (110) surface of TaIr$_{2}$B$_{2}$, we find extremely long surface Fermi arcs ($\sim$0.6 ${\textÅ}^{-1}$) located 1.4 meV below the Fermi level, which should be readily probed in experiment. Combined with the intrinsic superconductivity and the nontrivial bulk Fermi surfaces, $T$Ir$_{2}$B$_{2}$ may thus provide a very promising platform to explore the three-dimensional topological superconductivity.

cond-mat.supr-con↗

arXiv2020

Dynamic-K Recommendation with Personalized Decision Boundary

In this paper, we investigate the recommendation task in the most common scenario with implicit feedback (e.g., clicks, purchases). State-of-the-art methods in this direction usually cast the problem as to learn a personalized ranking on a set of items (e.g., webpages, products). The top-N results are then provided to users as recommendations, where the N is usually a fixed number pre-defined by the system according to some heuristic criteria (e.g., page size, screen size). There is one major assumption underlying this fixed-number recommendation scheme, i.e., there are always sufficient relevant items to users' preferences. Unfortunately, this assumption may not always hold in real-world scenarios. In some applications, there might be very limited candidate items to recommend, and some users may have very high relevance requirement in recommendation. In this way, even the top-1 ranked item may not be relevant to a user's preference. Therefore, we argue that it is critical to provide a dynamic-K recommendation, where the K should be different with respect to the candidate item set and the target user. We formulate this dynamic-K recommendation task as a joint learning problem with both ranking and classification objectives. The ranking objective is the same as existing methods, i.e., to create a ranking list of items according to users' interests. The classification objective is unique in this work, which aims to learn a personalized decision boundary to differentiate the relevant items from irrelevant items. Based on these ideas, we extend two state-of-the-art ranking-based recommendation methods, i.e., BPRMF and HRM, to the corresponding dynamic-K versions, namely DK-BPRMF and DK-HRM. Our experimental results on two datasets show that the dynamic-K models are more effective than the original fixed-N recommendation methods.

cs.IR↗

arXiv2020

"What Do You Mean by That?" A Parser-Independent Interactive Approach for Enhancing Text-to-SQL

In Natural Language Interfaces to Databases systems, the text-to-SQL technique allows users to query databases by using natural language questions. Though significant progress in this area has been made recently, most parsers may fall short when they are deployed in real systems. One main reason stems from the difficulty of fully understanding the users' natural language questions. In this paper, we include human in the loop and present a novel parser-independent interactive approach (PIIA) that interacts with users using multi-choice questions and can easily work with arbitrary parsers. Experiments were conducted on two cross-domain datasets, the WikiSQL and the more complex Spider, with five state-of-the-art parsers. These demonstrated that PIIA is capable of enhancing the text-to-SQL performance with limited interaction turns by using both simulation and human evaluation.

cs.CL↗

arXiv2020

Compositional Generalization by Learning Analytical Expressions

Compositional generalization is a basic and essential intellective capability of human beings, which allows us to recombine known parts readily. However, existing neural network based models have been proven to be extremely deficient in such a capability. Inspired by work in cognition which argues compositionality can be captured by variable slots with symbolic functions, we present a refreshing view that connects a memory-augmented neural model with analytical expressions, to achieve compositional generalization. Our model consists of two cooperative neural modules, Composer and Solver, fitting well with the cognitive argument while being able to be trained in an end-to-end manner via a hierarchical reinforcement learning algorithm. Experiments on the well-known benchmark SCAN demonstrate that our model seizes a great ability of compositional generalization, solving all challenges addressed by previous works with 100% accuracies.

cs.AI↗

arXiv2020

Robust Cross-View Gait Recognition with Evidence: A Discriminant Gait GAN (DiGGAN) Approach

Gait as a biometric trait has attracted much attention in many security and privacy applications such as identity recognition and authentication, during the last few decades. Because of its nature as a long-distance biometric trait, gait can be easily collected and used to identify individuals non-intrusively through CCTV cameras. However, it is very difficult to develop robust automated gait recognition systems, since gait may be affected by many covariate factors such as clothing, walking speed, camera view angle etc. Out of them, large view angle changes has been deemed as the most challenging factor as it can alter the overall gait appearance substantially. Existing works on gait recognition are far from enough to provide satisfying performances because of such view changes. Furthermore, very few works have considered evidences -- the demonstrable information revealing the reliabilities of decisions, which are regarded as important demands in machine learning-based recognition/authentication applications. To address these issues, in this paper we propose a Discriminant Gait Generative Adversarial Network, namely DiGGAN, which can effectively extract view-invariant features for cross-view gait recognition; and more importantly, to transfer gait images to different views -- serving as evidences and showing how the decisions have been made. Quantitative experiments have been conducted on the two most popular cross-view gait datasets, the OU-MVLP and CASIA-B, where the proposed DiGGAN has outperformed state-of-the-art methods. Qualitative analysis has also been provided and demonstrates the proposed DiGGAN's capability in providing evidences.

cs.CV↗