arXiv subjects

Xiaoli Wang

Publications and source records attributed to Xiaoli Wang.

53 records · Page 3Linked to original sources

arXiv2023

Search-Map-Search: A Frame Selection Paradigm for Action Recognition

Despite the success of deep learning in video understanding tasks, processing every frame in a video is computationally expensive and often unnecessary in real-time applications. Frame selection aims to extract the most informative and representative frames to help a model better understand video content. Existing frame selection methods either individually sample frames based on per-frame importance prediction, without considering interaction among frames, or adopt reinforcement learning agents to find representative frames in succession, which are costly to train and may lead to potential stability issues. To overcome the limitations of existing methods, we propose a Search-Map-Search learning paradigm which combines the advantages of heuristic search and supervised learning to select the best combination of frames from a video as one entity. By combining search with learning, the proposed method can better capture frame interactions while incurring a low inference overhead. Specifically, we first propose a hierarchical search method conducted on each training video to search for the optimal combination of frames with the lowest error on the downstream task. A feature mapping function is then learned to map the frames of a video to the representation of its target optimal frame combination. During inference, another search is performed on an unseen video to select a combination of frames whose feature representation is close to the projected feature representation. Extensive experiments based on several action recognition benchmarks demonstrate that our frame selection method effectively improves performance of action recognition models, and significantly outperforms a number of competitive baselines.

cs.CV↗

arXiv2023

LA3: Efficient Label-Aware AutoAugment

Automated augmentation is an emerging and effective technique to search for data augmentation policies to improve generalizability of deep neural network training. Most existing work focuses on constructing a unified policy applicable to all data samples in a given dataset, without considering sample or class variations. In this paper, we propose a novel two-stage data augmentation algorithm, named Label-Aware AutoAugment (LA3), which takes advantage of the label information, and learns augmentation policies separately for samples of different labels. LA3 consists of two learning stages, where in the first stage, individual augmentation methods are evaluated and ranked for each label via Bayesian Optimization aided by a neural predictor, which allows us to identify effective augmentation techniques for each label under a low search cost. And in the second stage, a composite augmentation policy is constructed out of a selection of effective as well as complementary augmentations, which produces significant performance boost and can be easily deployed in typical model training. Extensive experiments demonstrate that LA3 achieves excellent performance matching or surpassing existing methods on CIFAR-10 and CIFAR-100, and achieves a new state-of-the-art ImageNet accuracy of 79.97% on ResNet-50 among auto-augmentation methods, while maintaining a low computational cost.

cs.CV↗

arXiv2023

WR-ONE2SET: Towards Well-Calibrated Keyphrase Generation

Keyphrase generation aims to automatically generate short phrases summarizing an input document. The recently emerged ONE2SET paradigm (Ye et al., 2021) generates keyphrases as a set and has achieved competitive performance. Nevertheless, we observe serious calibration errors outputted by ONE2SET, especially in the over-estimation of $\varnothing$ token (means "no corresponding keyphrase"). In this paper, we deeply analyze this limitation and identify two main reasons behind: 1) the parallel generation has to introduce excessive $\varnothing$ as padding tokens into training instances; and 2) the training mechanism assigning target to each slot is unstable and further aggravates the $\varnothing$ token over-estimation. To make the model well-calibrated, we propose WR-ONE2SET which extends ONE2SET with an adaptive instance-level cost Weighting strategy and a target Re-assignment mechanism. The former dynamically penalizes the over-estimated slots for different instances thus smoothing the uneven training distribution. The latter refines the original inappropriate assignment and reduces the supervisory signals of over-estimated slots. Experimental results on commonly-used datasets demonstrate the effectiveness and generality of our proposed paradigm.

cs.CL↗

arXiv2021

Security analysis method for practical quantum key distribution with arbitrary encoding schemes

Quantum key distribution (QKD) gradually has become a crucial element of practical secure communication. In different scenarios, the security analysis of genuine QKD systems is complicated. A universal secret key rate calculation method, used for realistic factors such as multiple degrees of freedom encoding, asymmetric protocol structures, equipment flaws, environmental noise, and so on, is still lacking. Based on the correlations of statistical data, we propose a security analysis method without restriction on encoding schemes. This method makes a trade-off between applicability and accuracy, which can effectively analyze various existing QKD systems. We illustrate its ability by analyzing source flaws and a high-dimensional asymmetric protocol. Results imply that our method can give tighter bounds than the Gottesman-Lo-Lütkenhaus-Preskill (GLLP) analysis and is beneficial to analyze protocols with complex encoding structures. Our work has the potential to become a reference standard for the security analysis of practical QKD.

quant-ph↗

arXiv2021

Verdi: Quality Estimation and Error Detection for Bilingual Corpora

Translation Quality Estimation is critical to reducing post-editing efforts in machine translation and to cross-lingual corpus cleaning. As a research problem, quality estimation (QE) aims to directly estimate the quality of translation in a given pair of source and target sentences, and highlight the words that need corrections, without referencing to golden translations. In this paper, we propose Verdi, a novel framework for word-level and sentence-level post-editing effort estimation for bilingual corpora. Verdi adopts two word predictors to enable diverse features to be extracted from a pair of sentences for subsequent quality estimation, including a transformer-based neural machine translation (NMT) model and a pre-trained cross-lingual language model (XLM). We exploit the symmetric nature of bilingual corpora and apply model-level dual learning in the NMT predictor, which handles a primal task and a dual task simultaneously with weight sharing, leading to stronger context prediction ability than single-direction NMT models. By taking advantage of the dual learning scheme, we further design a novel feature to directly encode the translated target information without relying on the source context. Extensive experiments conducted on WMT20 QE tasks demonstrate that our method beats the winner of the competition and outperforms other baseline methods by a great margin. We further use the sentence-level scores provided by Verdi to clean a parallel corpus and observe benefits on both model performance and training efficiency.

cs.CL↗

arXiv2021

A Multi-Scale Method for Distributed Convex Optimization with Constraints

This paper proposes a multi-scale method to design a continuous-time distributed algorithm for constrained convex optimization problems by using multi-agents with Markov switched network dynamics and noisy inter-agent communications. Unlike most previous work which mainly puts emphasis on dealing with fixed network topology, this paper tackles the challenging problem of investigating the joint effects of stochastic networks and the inter-agent communication noises on the distributed optimization dynamics, which has not been systemically studied in the past literature. Also, in sharp contrast to previous work in constrained optimization, we depart from the use of projected gradient flow which is non-smooth and hard to analyze; instead, we design a smooth optimization dynamics which leads to easier convergence analysis and more efficient numerical simulations. Moreover, the multi-scale method presented in this paper generalizes previously known distributed convex optimization algorithms from the fixed network topology to the switching case and the stochastic averaging obtained in this paper is a generalization of the existing deterministic averaging.

math.OC↗

arXiv2020

Reinforced Curriculum Learning on Pre-trained Neural Machine Translation Models

The competitive performance of neural machine translation (NMT) critically relies on large amounts of training data. However, acquiring high-quality translation pairs requires expert knowledge and is costly. Therefore, how to best utilize a given dataset of samples with diverse quality and characteristics becomes an important yet understudied question in NMT. Curriculum learning methods have been introduced to NMT to optimize a model's performance by prescribing the data input order, based on heuristics such as the assessment of noise and difficulty levels. However, existing methods require training from scratch, while in practice most NMT models are pre-trained on big data already. Moreover, as heuristics, they do not generalize well. In this paper, we aim to learn a curriculum for improving a pre-trained NMT model by re-selecting influential data samples from the original training set and formulate this task as a reinforcement learning problem. Specifically, we propose a data selection framework based on Deterministic Actor-Critic, in which a critic network predicts the expected change of model performance due to a certain sample, while an actor network learns to select the best sample out of a random batch of samples presented to it. Experiments on several translation datasets show that our method can further improve the performance of NMT when original batch training reaches its ceiling, without using additional new training data, and significantly outperforms several strong baseline methods.

cs.CL↗

arXiv2020

Joint Source and Relay Design for MIMO Relaying Broadcast Channels

In this letter, we address the optimal source and relay matrices design for the multiple-input multiple-output~(MIMO) relaying broadcast channels~(BC) with direct links~(DLs) based on weighted sum-rate criterion.This problem is nonlinear nonconvex and its optimal solution remains open. To develop an efficient way to solve this problem,we first set up an equivalent problem, and solve it by decoupling it into two tractable subproblems. Finally, we propose a general linear iterative design algorithm based on alternative optimization. This iterative design algorithm is convergent since the solution of each subproblem is optimal and unique. The advantage of the proposed iterative design scheme is demonstrated by numerical experiments.

eess.SP↗

arXiv2019

A wide star-black-hole binary system from radial-velocity measurements

All stellar mass black holes have hitherto been identified by X-rays emitted by gas that is accreting onto the black hole from a companion star. These systems are all binaries with black holes below 30 M$_{\odot}$$^{1-4}$. Theory predicts, however, that X-ray emitting systems form a minority of the total population of star-black hole binaries$^{5,6}$. When the black hole is not accreting gas, it can be found through radial velocity measurements of the motion of the companion star. Here we report radial velocity measurements of a Galactic star, LB-1, which is a B-type star, taken over two years. We find that the motion of the B-star and an accompanying H$α$ emission line require the presence of a dark companion with a mass of $68^{+11}_{-13}$ M$_{\odot}$, which can only be a black hole. The long orbital period of 78.9 days shows that this is a wide binary system. The gravitational wave experiments have detected similarly massive black holes$^{7,8}$, but forming such massive ones in a high-metallicity environment would be extremely challenging to current stellar evolution theories$^{9-11}$.

astro-ph.SR↗

arXiv2019

Reverse chemistry of iron in the deep Earth

In this work, we demonstrate a remarkable change of chemical trend of Iron under high pressure that is of great importance for understanding the distribution of elements in the Earth's mantle and core. Using first principles crystal structure search method, we conduct a systematic study of the propensity of p block elements to chemically bind with iron under high pressures ranging from ambient conditions to that of Earth's core. We show that under increasing pressure, iron tends to reverse its chemical nature, changing from an electron donor (reductant) to an electron acceptor, and oxidizes p-block elements in many compounds. Such reverse chemistry has a significant impact on the stoichiometries, bond types and strengths, structures and properties of iron compounds under deep planetary conditions.

physics.chem-ph↗

arXiv2018

Towards practical high-speed high dimensional quantum key distribution using partial mutual unbiased basis of photon's orbital angular momentum

Quantum Key Distribution (QKD) guarantees the security of communication with quantum physics. Most of widely adopted QKD protocols currently encode the key information with binary signal format---qubit, such as the polarization states. Therefore the transmitted information efficiency of the quantum key is intrinsically upper bounded by 1 bit per photon. High dimensional quantum system is a potential candidate for increasing the capacity of single photon. However, due to the difficulty in manipulating and measuring high dimensional quantum systems, the experimental high dimensional QKD is still at its infancy. Here we propose a sort of practical high-speed high dimensional QKD using partial mutual unbiased basis (PMUB) of photon's orbital angular momentum (OAM). Different from the previous OAM encoding, the high dimensional Hilbert space we used is expanded by the OAM states with same mode order, which can be extended to considerably high dimensions and implemented under current state of the art. Because all the OAM states are in the same mode order, the coherence will be well kept after long-distance propagation, and the detection can be achieved by using passive linear optical elements with very high speed. We show that our protocol has high key generation rate and analyze the anti-noise ability under atmospheric turbulence. Furthermore, the security of our protocol based on PMUB is rigorously proved. Our protocol paves a brand new way for the application of photon's OAM in high dimensional QKD field, which can be a breakthrough for high efficiency quantum communications.

quant-ph↗

arXiv2015

Exploring genetic variation in the tomato (Solanum section Lycopersicon) clade by whole-genome sequencing

Genetic variation in the tomato clade was explored by sequencing a selection of 84 tomato accessions and related wild species representative for the Lycopersicon, Arcanum, Eriopersicon, and Neolycopersicon groups. We present a reconstruction of three new reference genomes in support of our comparative genome analyses. Sequence diversity in commercial breeding lines appears extremely low, indicating the dramatic genetic erosion of crop tomatoes. This is reflected by the SNP count in wild species which can exceed 10 million i.e. 20 fold higher than in crop accessions. Comparative sequence alignment reveals group, species, and accession specific polymorphisms, which explain characteristic fruit traits and growth habits in tomato accessions. Using gene models from the annotated Heinz reference genome, we observe a bias in dN/dS ratio in fruit and growth diversification genes compared to a random set of genes, which probably is the result of a positive selection. We detected highly divergent segments in wild S. lycopersicum species, and footprints of introgressions in crop accessions originating from a common donor accession. Phylogenetic relationships of fruit diversification and growth specific genes from crop accessions show incomplete resolution and are dependent on the introgression donor. In contrast, whole genome SNP information has sufficient power to resolve the phylogenetic placement of each accession in the four main groups in the Lycopersicon clade using Maximum Likelihood analyses. Phylogenetic relationships appear correlated with habitat and mating type and point to the occurrence of geographical races within these groups and thus are of practical importance for introgressive hybridization breeding. Our study illustrates the need for multiple reference genomes in support of tomato comparative genomics and Solanum genome evolution studies.

q-bio.GN↗

arXiv2014

An embryo of protocell membrane: The capsule of graphene oxide

Many signs indicate that the graphene could widely occur on the early Earth. Here, we report a new theory that graphene might be an embryo of protocell membrane, and found several evidences. Firstly, the graphene oxide and phospholipid-graphene oxide composite would curl into capsules in strongly acidic saturated solution of Pb(NO3)2 at low temperature, providing a protective space for biochemical reactions. Secondly, L-animi acids exhibit higher reactivity than D-animi acids for graphene oxides in favor of the formation of left-handed proteins. Thirdly, monolayer graphene with nanopores prepared by unfocused 84Kr25+ has high selectivity for permeation of the monovalent metal ions (Rb+ > K+ > Cs+ > Na+ > Li+), but does not allow Cl- through, which could be attributed to the ion exchange of oxygen-containing groups on the rim of nanopores. It is similar to K+ channels, which would cause efflux of some ions from capsule of graphene oxides with the decrease of pH in the primitive ocean, creating a suitable inner condition for the origin of life. Consequently, the strongly acidic, high salinity and strong radiation as well as temperature changes in the early Earth, regarded as negative factors, would be indispensable for the origin of protocell. In short, graphene bred life, but digested gradually by the evolution.

q-bio.PE↗

arXiv2012

Efficiency Resource Allocation for Device-to-Device Underlay Communication Systems: A Reverse Iterative Combinatorial Auction Based Approach

Peer-to-peer communication has been recently considered as a popular issue for local area services. An innovative resource allocation scheme is proposed to improve the performance of mobile peer-to-peer, i.e., device-to-device (D2D), communications as an underlay in the downlink (DL) cellular networks. To optimize the system sum rate over the resource sharing of both D2D and cellular modes, we introduce a reverse iterative combinatorial auction as the allocation mechanism. In the auction, all the spectrum resources are considered as a set of resource units, which as bidders compete to obtain business while the packages of the D2D pairs are auctioned off as goods in each auction round. We first formulate the valuation of each resource unit, as a basis of the proposed auction. And then a detailed non-monotonic descending price auction algorithm is explained depending on the utility function that accounts for the channel gain from D2D and the costs for the system. Further, we prove that the proposed auction-based scheme is cheat-proof, and converges in a finite number of iteration rounds. We explain non-monotonicity in the price update process and show lower complexity compared to a traditional combinatorial allocation. The simulation results demonstrate that the algorithm efficiently leads to a good performance on the system sum rate.

cs.GT↗

arXiv2012

Leader-Following Consensus of Multiple Linear Systems Under Switching Topologies: An Averaging Method

The leader-following consensus of multiple linear time invariant (LTI) systems under switching topology is considered. The leader-following consensus problem consists of designing for each agent a distributed protocol to make all agents track a leader vehicle, which has the same LTI dynamics as the agents. The interaction topology describing the information exchange of these agents is time-varying. An averaging method is proposed. Unlike the existing results in the literatures which assume the LTI agents to be neutrally stable, we relax this condition, only making assumption that the LTI agents are stablizable and detectable. Observer-based leader-following consensus is also considered.

math.OC↗

arXiv2012

Diamondoid Structure of Polymeric Nitrogen at High Pressures

High-pressure polymeric structures of nitrogen have attracted great attention owing to their potential application as high-energy-density materials. We report the density functional structural prediction of the unexpected stabilization of a diamondoid (or N10-cage) structure of polymeric nitrogen at high pressures. The structure adopts a highly symmetric body-centered cubic form with lattice sites occupied by N10 tetracyclic cages, each of which consists of 10 atoms and is covalently bonded with its six next-nearest N10 cages. The prediction of this diamondoid structure rules out the earlier proposed helical tunnel phase and demonstrates the high-order nature of polymeric nitrogen at extreme high pressures. Diamondoid nitrogen is a wide-gap insulator and energetically more favorable than the experimental cubic gauche and previously predicted layered Pba2 phases above 263 GPa, a pressure which is accessible to high pressure experiment.

cond-mat.mtrl-sci↗

arXiv2012

High pressures make Hg a transition metal in a thermodynamically stable solid

The appropriateness of including Hg among the transition metals has been debated for a long time. Although the synthesis of HgF$_{4}$ molecules in gas phase was reported before, the molecules show strong instabilities and dissociate. Therefore, the transition metal propensity of Hg remains an open question. Here, we propose that high pressure provides a controllable method for preparing unusual oxidation states of matter. Using an advanced structure search method based on first-principles electronic structure calculations, we predict that under high pressures, Hg can transfer the electrons in its outmost $d$ shell to F atoms, thereby acting as a transition metal. Oxidation of Hg to the +4 state yielded thermodynamically stable molecular crystals consisting of HgF$_{4}$ planar molecules, a typical geometry for $d^{8}$ metal centers.

cond-mat.mtrl-sci↗