arXiv subjects

Xin Liu

Publications and source records attributed to Xin Liu.

At least 775 records · Page 43Linked to original sources

arXiv2021

KGR^4: Retrieval, Retrospect, Refine and Rethink for Commonsense Generation

Generative commonsense reasoning requires machines to generate sentences describing an everyday scenario given several concepts, which has attracted much attention recently. However, existing models cannot perform as well as humans, since sentences they produce are often implausible and grammatically incorrect. In this paper, inspired by the process of humans creating sentences, we propose a novel Knowledge-enhanced Commonsense Generation framework, termed KGR^4, consisting of four stages: Retrieval, Retrospect, Refine, Rethink. Under this framework, we first perform retrieval to search for relevant sentences from external corpus as the prototypes. Then, we train the generator that either edits or copies these prototypes to generate candidate sentences, of which potential errors will be fixed by an autoencoder-based refiner. Finally, we select the output sentence from candidate sentences produced by generators with different hyper-parameters. Experimental results and in-depth analysis on the CommonGen benchmark strongly demonstrate the effectiveness of our framework. Particularly, KGR^4 obtains 33.56 SPICE points in the official leaderboard, outperforming the previously-reported best result by 2.49 SPICE points and achieving state-of-the-art performance.

cs.CL↗

arXiv2021

Varstrometry for Off-nucleus and Dual sub-Kpc AGN (VODKA): Hubble Space Telescope Discovers Double Quasars

Dual supermassive black holes (SMBHs) at $\sim$kpc scales are the progenitor population of SMBH mergers and play an important role in understanding the pairing and dynamical evolution of massive black holes in galaxy mergers. Because of the stringent resolution requirement and the apparent rareness of these small-separation pairs, there are scarce observational constraints on this population, with few confirmed dual SMBHs at $<10$kpc separations at $z>1$. Here we present results from a pilot search for kpc-scale dual quasars selected with Gaia Data release 2 (DR2) astrometry and followed up with Hubble Space Telescope (HST) Wide Field Camera 3 dual-band (F475W and F814W) snapshot imaging. Our targets are quasars primarily selected with the varstrometry technique, i.e., light centroid jitter caused by asynchronous variability from both members in an unresolved quasar pair, supplemented by sub-arcsec pairs already resolved by Gaia DR2. We find an overall high fraction of HST-resolved pairs among the varstrometry-selected quasars (unresolved in Gaia DR2), $\sim 30-50\%$, increasing toward high redshift ($\sim 60-80\%$ at $z>1.5$). We discuss the nature of the 43 resolved sub-arcsec pairs based on HST and supplementary data. A substantial fraction ($\sim 40\%$) of these pairs are likely physical quasar pairs or gravitationally lensed quasars. We also discover a triple quasar candidate and a quadruply lensed quasar, which is among the smallest-separation quadruple lenses. These results provide important guidelines to improve varstrometry selection and follow-up confirmation of $\sim$kpc-scale dual SMBHs at high redshift.

astro-ph.HE↗

arXiv2021

Improving the Transferability of Adversarial Examples with Resized-Diverse-Inputs, Diversity-Ensemble and Region Fitting

We introduce a three stage pipeline: resized-diverse-inputs (RDIM), diversity-ensemble (DEM) and region fitting, that work together to generate transferable adversarial examples. We first explore the internal relationship between existing attacks, and propose RDIM that is capable of exploiting this relationship. Then we propose DEM, the multi-scale version of RDIM, to generate multi-scale gradients. After the first two steps we transform value fitting into region fitting across iterations. RDIM and region fitting do not require extra running time and these three steps can be well integrated into other attacks. Our best attack fools six black-box defenses with a 93% success rate on average, which is higher than the state-of-the-art gradient-based attacks. Besides, we rethink existing attacks rather than simply stacking new methods on the old ones to get better performance. It is expected that our findings will serve as the beginning of exploring the internal relationship between attack methods. Codes are available at https://github.com/278287847/DEM.

cs.CV↗

arXiv2021

Natural Image Reconstruction from fMRI using Deep Learning: A Survey

With the advent of brain imaging techniques and machine learning tools, much effort has been devoted to building computational models to capture the encoding of visual information in the human brain. One of the most challenging brain decoding tasks is the accurate reconstruction of the perceived natural images from brain activities measured by functional magnetic resonance imaging (fMRI). In this work, we survey the most recent deep learning methods for natural image reconstruction from fMRI. We examine these methods in terms of architectural design, benchmark datasets, and evaluation metrics and present a fair performance evaluation across standardized evaluation metrics. Finally, we discuss the strengths and limitations of existing studies and present potential future directions.

cs.CV↗

arXiv2021

Consistency Regularization for Deep Face Anti-Spoofing

Face anti-spoofing (FAS) plays a crucial role in securing face recognition systems. Empirically, given an image, a model with more consistent output on different views of this image usually performs better, as shown in Fig.1. Motivated by this exciting observation, we conjecture that encouraging feature consistency of different views may be a promising way to boost FAS models. In this paper, we explore this way thoroughly by enhancing both Embedding-level and Prediction-level Consistency Regularization (EPCR) in FAS. Specifically, at the embedding-level, we design a dense similarity loss to maximize the similarities between all positions of two intermediate feature maps in a self-supervised fashion; while at the prediction-level, we optimize the mean square error between the predictions of two views. Notably, our EPCR is free of annotations and can directly integrate into semi-supervised learning schemes. Considering different application scenarios, we further design five diverse semi-supervised protocols to measure semi-supervised FAS techniques. We conduct extensive experiments to show that EPCR can significantly improve the performance of several supervised and semi-supervised tasks on benchmark datasets. The codes and protocols will be released at https://github.com/clks-wzz/EPCR.

cs.CV↗

arXiv2021

Braiding higher-order Majorana corner states through their spin degree of freedom

In this work, we study the spin texture of a class of higher-order topological superconductors (HOTSC) and show how it can be used to detect and braid Majorana corner modes (MCMs). This class of HOTSC is composed of two-dimensional topological insulators with s-wave superconductivity and in-plane magnetic fields, which offers advantages in experimental implementation. The spin polarization of the MCMs in this class is perpendicular with the applied magnetic field direction and is opposite on intrinsic orbitals, resulting in an overall ferrimagnetic spin texture. As a result, we find that the spin-selective Andreev reflection can be observed in a transverse instead of parallel direction to the applied magnetic field. Meanwhile, this spin texture leads to the gate-tunable $4π$ periodic $ϕ_0$ Josephson current that performs qualitatively different behavior from the topologically trivial $ϕ_0$-junction under rotating the in-plane magnetic field. Meanwhile, the existence of the MCMs in this class does not depend on the in-plane magnetic field direction. This gives rise to great advantage in constructing all electronically controlled Majorana network for braiding, which is confirmed through our numerical simulation. We thus provide a comprehensive scheme for probing non-Abelian statistics in this class of HOTSCs.

cond-mat.mes-hall↗

arXiv2021

Redefining the Quantum Supremacy Baseline With a New Generation Sunway Supercomputer

A major milestone in the era of noisy intermediate scale quantum computers is \textit{quantum supremacy} [Nature \textbf{574}, 505 (2019)] claimed on the Sycamore quantum processor of $53$ qubits, which can perform a random circuit sampling task within $200$ seconds while the same task is estimated to require a runtime of $10,000$ years on Summit. This record has been renewed with two recent experiments on the Zuchongzhi $2.0$ ($56$ qubits) and Zuchongzhi $2.1$ ($60$ qubits) quantum processors. On the other front of quantum supremacy comparison, there has also been continuous improvements on both the classical simulation algorithm as well as the underlying hardware. And a fair justification of the computational advantages for those quantum supremacy experiments would require to practically simulate the same problems on current top supercomputers, which is still in lack. Here we report the full-scale simulations of these problems on new generation Sunway supercomputer, based on a customized tensor network contraction algorithm. Our benchmark shows that the most challenging sampling task performed on Sycamore can be accomplished within $1$ week, thus collapsing the quantum supremacy claim of Sycamore. Additionally, we show that the XEB fidelities of the \textit{quantum supremacy circuits} with up to $14$ cycles can be verified in minutes, which also provides strong consistency check for quantum supremacy experiments. Our results redefine quantum supremacy baseline using the new generation Sunway supercomputer.

quant-ph↗

arXiv2021

CLARA: A Constrained Reinforcement Learning Based Resource Allocation Framework for Network Slicing

As mobile networks proliferate, we are experiencing a strong diversification of services, which requires greater flexibility from the existing network. Network slicing is proposed as a promising solution for resource utilization in 5G and future networks to address this dire need. In network slicing, dynamic resource orchestration and network slice management are crucial for maximizing resource utilization. Unfortunately, this process is too complex for traditional approaches to be effective due to a lack of accurate models and dynamic hidden structures. We formulate the problem as a Constrained Markov Decision Process (CMDP) without knowing models and hidden structures. Additionally, we propose to solve the problem using CLARA, a Constrained reinforcement LeArning based Resource Allocation algorithm. In particular, we analyze cumulative and instantaneous constraints using adaptive interior-point policy optimization and projection layer, respectively. Evaluations show that CLARA clearly outperforms baselines in resource allocation with service demand guarantees.

cs.NI↗

arXiv2021

Simplifying approach to Node Classification in Graph Neural Networks

Graph Neural Networks have become one of the indispensable tools to learn from graph-structured data, and their usefulness has been shown in wide variety of tasks. In recent years, there have been tremendous improvements in architecture design, resulting in better performance on various prediction tasks. In general, these neural architectures combine node feature aggregation and feature transformation using learnable weight matrix in the same layer. This makes it challenging to analyze the importance of node features aggregated from various hops and the expressiveness of the neural network layers. As different graph datasets show varying levels of homophily and heterophily in features and class label distribution, it becomes essential to understand which features are important for the prediction tasks without any prior information. In this work, we decouple the node feature aggregation step and depth of graph neural network, and empirically analyze how different aggregated features play a role in prediction performance. We show that not all features generated via aggregation steps are useful, and often using these less informative features can be detrimental to the performance of the GNN model. Through our experiments, we show that learning certain subsets of these features can lead to better performance on wide variety of datasets. We propose to use softmax as a regularizer and "soft-selector" of features aggregated from neighbors at different hop distances; and L2-Normalization over GNN layers. Combining these techniques, we present a simple and shallow model, Feature Selection Graph Neural Network (FSGNN), and show empirically that the proposed model achieves comparable or even higher accuracy than state-of-the-art GNN models in nine benchmark datasets for the node classification task, with remarkable improvements up to 51.1%.

stat.ML↗

arXiv2021

An Efficient Pessimistic-Optimistic Algorithm for Stochastic Linear Bandits with General Constraints

This paper considers stochastic linear bandits with general nonlinear constraints. The objective is to maximize the expected cumulative reward over horizon $T$ subject to a set of constraints in each round $τ\leq T$. We propose a pessimistic-optimistic algorithm for this problem, which is efficient in two aspects. First, the algorithm yields $\tilde{\cal O}\left(\left(\frac{K^{0.75}}δ+d\right)\sqrtτ\right)$ (pseudo) regret in round $τ\leq T,$ where $K$ is the number of constraints, $d$ is the dimension of the reward feature space, and $δ$ is a Slater's constant; and zero constraint violation in any round $τ>τ',$ where $τ'$ is independent of horizon $T.$ Second, the algorithm is computationally efficient. Our algorithm is based on the primal-dual approach in optimization and includes two components. The primal component is similar to unconstrained stochastic linear bandits (our algorithm uses the linear upper confidence bound algorithm (LinUCB)). The computational complexity of the dual component depends on the number of constraints, but is independent of the sizes of the contextual space, the action space, and the feature space. Thus, the overall computational complexity of our algorithm is similar to that of the linear UCB for unconstrained stochastic linear bandits.

cs.LG↗

arXiv2021

Charmless $B_s\to V S$ Decays in PQCD Approach

In this work, we investigate the $B_s\to V S$ decays in the perturbative QCD approach, where $V$ and $S$ denote the vector meson and scalar meson respectively. Based on the two-quark structure, considering two different scenarios for describing the scalar mesons, we calculate the branching fractions and the direct $CP$ asymmetries of all $B_s\to VS$ decays. Most branching fractions are predicted to be at $10^{-7}$ to $10^{-5}$, which could be measured in the LHCb and Belle-II experiments, especially for these color-allowed $B_s\to κ(800)(K_0^*(1430))K^*$ decays. It is found that the branching fractions of $B_s\to K_0^{*0}(1430)\bar{K}^{*0}$ and $B_s\to K_0^{*+}(1430)\bar{K}^{*-}$ are very sensitive to the scenarios, which can be used to determine whether $K_0^{*0}(1430)$ belongs to the ground state or the first excited state, if the data were available. We also note that some decays have large direct $CP$ asymmetries, some of which are also sensitive to the scenarios, such as the $B_s \to a_0^+(1450)K^{*-}$ and the $B_s\to f_0(1500) K^{*0}$ decays. Since the experimental measurements of $B_s\to VS$ decays are on the way, combined with the available data in the future, we expect the theoretical predictions will shed light on the structure of the scalar mesons.

hep-ph↗

arXiv2021

A Provably-Efficient Model-Free Algorithm for Constrained Markov Decision Processes

This paper presents the first model-free, simulator-free reinforcement learning algorithm for Constrained Markov Decision Processes (CMDPs) with sublinear regret and zero constraint violation. The algorithm is named Triple-Q because it includes three key components: a Q-function (also called action-value function) for the cumulative reward, a Q-function for the cumulative utility for the constraint, and a virtual-Queue that (over)-estimates the cumulative constraint violation. Under Triple-Q, at each step, an action is chosen based on the pseudo-Q-value that is a combination of the three "Q" values. The algorithm updates the reward and utility Q-values with learning rates that depend on the visit counts to the corresponding (state, action) pairs and are periodically reset. In the episodic CMDP setting, Triple-Q achieves $\tilde{\cal O}\left(\frac{1 }δH^4 S^{\frac{1}{2}}A^{\frac{1}{2}}K^{\frac{4}{5}} \right)$ regret, where $K$ is the total number of episodes, $H$ is the number of steps in each episode, $S$ is the number of states, $A$ is the number of actions, and $δ$ is Slater's constant. Furthermore, Triple-Q guarantees zero constraint violation, both on expectation and with a high probability, when $K$ is sufficiently large. Finally, the computational complexity of Triple-Q is similar to SARSA for unconstrained MDPs and is computationally efficient.

cs.LG↗

arXiv2021

0-$π$ qubit in one Josephson junction

Quantum states are usually fragile which makes quantum computation being not as stable as classical computation. Quantum correction codes can protect quantum states but need a large number of physical qubits to code a single logic qubit. Alternatively, the protection at the hardware level has been recently developed to maintain the coherence of the quantum information by using symmetries. However, it generally has to pay the expense of increasing the complexity of the quantum devices. In this work, we show that the protection at the hardware level can be approached without increasing the complexity of the devices. The interplay between the spin-orbit coupling and the Zeeman splitting in the semiconductor allows us to tune the Josephson coupling in terms of the spin degree of freedom of Cooper pairs, the hallmark of the superconducting spintronics. This leads to the implementation of the parity-protected 0-$π$ superconducting qubit with only one highly transparent superconductor-semiconductor Josephson junction, which makes our proposal immune from the various fabrication imperfections.

cond-mat.mes-hall↗

arXiv2021

A Global Optimization Approach for Multi-Marginal Optimal Transport Problems with Coulomb Cost

In this work, we construct a novel numerical method for solving the multi-marginal optimal transport problems with Coulomb cost. This type of optimal transport problems arises in quantum physics and plays an important role in understanding the strongly correlated quantum systems. With a Monge-like ansatz, the orginal high-dimensional problems are transferred into mathematical programmings with generalized complementarity constraints, and thus the curse of dimensionality is surmounted. However, the latter ones are themselves hard to deal with from both theoretical and practical perspective. Moreover in the presence of nonconvexity, brute-force searching for global solutions becomes prohibitive as the problem size grows large. To this end, we propose a global optimization approach for solving the nonconvex optimization problems, by exploiting an efficient proximal block coordinate descent local solver and an initialization subroutine based on hierarchical grid refinements. We provide numerical simulations on some typical physical systems to show the efficiency of our approach. The results match well with both theoretical predictions and physical intuitions, and give the first visualization of optimal transport maps for some two dimensional systems.

math.OC↗

arXiv2021

Synthetic Data for Multi-Parameter Camera-Based Physiological Sensing

Synthetic data is a powerful tool in training data hungry deep learning algorithms. However, to date, camera-based physiological sensing has not taken full advantage of these techniques. In this work, we leverage a high-fidelity synthetics pipeline for generating videos of faces with faithful blood flow and breathing patterns. We present systematic experiments showing how physiologically-grounded synthetic data can be used in training camera-based multi-parameter cardiopulmonary sensing. We provide empirical evidence that heart and breathing rate measurement accuracy increases with the number of synthetic avatars in the training set. Furthermore, training with avatars with darker skin types leads to better overall performance than training with avatars with lighter skin types. Finally, we discuss the opportunities that synthetics present in the domain of camera-based physiological sensing and limitations that need to be overcome.

cs.CV↗

arXiv2021

Investigating the Accretion Nature of Binary Supermassive Black Hole Candidate SDSS J025214.67-002813.7

We present results on a multi-wavelength analysis of SDSS J025214.67-002813.7, a system which has been previously classified as a binary AGN candidate based on periodic signals detected in the optical light curves. We use available radio-X-ray observations of the system to investigate the true accretion nature. Analyzing new observations from XMM-Newton and NuSTAR, we characterize the X-ray emission and search for evidence of circumbinary accretion. Although the 0.5-10 keV spectrum shows evidence of an additional soft emission component, possibly due to extended emission from hot nuclear gas, we find the spectral shape consistent with a single AGN. Compiling a full multi-wavelength SED, we also search for signs of circumbinary accretion, such as a "notch" in the continuum due to the presence of minidisks. We find that the radio-optical emission agrees with the SED of a standard, radio-quiet, AGN, however there is a large deficit in emission blueward of ~1400 A. Although this deficit in emission can plausibly be attributed to a binary AGN system, we find that the SED of SDSS J0252-0028 is better explained by emission from a reddened, single AGN. However, future studies on the expected hard X-ray emission associated with binary AGN (especially in the unequal-mass regime), will allow for more rigorous analyses of the binary AGN hypothesis.

astro-ph.HE↗

arXiv2021

Annotation-efficient deep learning for automatic medical image segmentation

Automatic medical image segmentation plays a critical role in scientific research and medical care. Existing high-performance deep learning methods typically rely on large training datasets with high-quality manual annotations, which are difficult to obtain in many clinical applications. Here, we introduce Annotation-effIcient Deep lEarning (AIDE), an open-source framework to handle imperfect training datasets. Methodological analyses and empirical evaluations are conducted, and we demonstrate that AIDE surpasses conventional fully-supervised models by presenting better performance on open datasets possessing scarce or noisy annotations. We further test AIDE in a real-life case study for breast tumor segmentation. Three datasets containing 11,852 breast images from three medical centers are employed, and AIDE, utilizing 10% training annotations, consistently produces segmentation maps comparable to those generated by fully-supervised counterparts or provided by independent radiologists. The 10-fold enhanced efficiency in utilizing expert labels has the potential to promote a wide range of biomedical applications.

eess.IV↗

arXiv2021

Cross-lingual Transfer for Text Classification with Dictionary-based Heterogeneous Graph

In cross-lingual text classification, it is required that task-specific training data in high-resource source languages are available, where the task is identical to that of a low-resource target language. However, collecting such training data can be infeasible because of the labeling cost, task characteristics, and privacy concerns. This paper proposes an alternative solution that uses only task-independent word embeddings of high-resource languages and bilingual dictionaries. First, we construct a dictionary-based heterogeneous graph (DHG) from bilingual dictionaries. This opens the possibility to use graph neural networks for cross-lingual transfer. The remaining challenge is the heterogeneity of DHG because multiple languages are considered. To address this challenge, we propose dictionary-based heterogeneous graph neural network (DHGNet) that effectively handles the heterogeneity of DHG by two-step aggregations, which are word-level and language-level aggregations. Experimental results demonstrate that our method outperforms pretrained models even though it does not access to large corpora. Furthermore, it can perform well even though dictionaries contain many incorrect translations. Its robustness allows the usage of a wider range of dictionaries such as an automatically constructed dictionary and crowdsourced dictionary, which are convenient for real-world applications.

cs.CL↗