Searcharxiv⌕ Search

arXiv subjects

Peng Xu

Publications and source records attributed to Peng Xu.

At least 361 records · Page 20Linked to original sources
arXiv2019

Clickbait? Sensational Headline Generation with Auto-tuned Reinforcement Learning

Sensational headlines are headlines that capture people's attention and generate reader interest. Conventional abstractive headline generation methods, unlike human writers, do not optimize for maximal reader attention. In this paper, we propose a model that generates sensational headlines without labeled data. We first train a sensationalism scorer by classifying online headlines with many comments ("clickbait") against a baseline of headlines generated from a summarization model. The score from the sensationalism scorer is used as the reward for a reinforcement learner. However, maximizing the noisy sensationalism reward will generate unnatural phrases instead of sensational headlines. To effectively leverage this noisy reward, we propose a novel loss function, Auto-tuned Reinforcement Learning (ARL), to dynamically balance reinforcement learning (RL) with maximum likelihood estimation (MLE). Human evaluation shows that 60.8% of samples generated by our model are sensational, which is significantly better than the Pointer-Gen baseline and other RL models.

cs.CL↗
arXiv2019

MoEL: Mixture of Empathetic Listeners

Previous research on empathetic dialogue systems has mostly focused on generating responses given certain emotions. However, being empathetic not only requires the ability of generating emotional responses, but more importantly, requires the understanding of user emotions and replying appropriately. In this paper, we propose a novel end-to-end approach for modeling empathy in dialogue systems: Mixture of Empathetic Listeners (MoEL). Our model first captures the user emotions and outputs an emotion distribution. Based on this, MoEL will softly combine the output states of the appropriate Listener(s), which are each optimized to react to certain emotions, and generate an empathetic response. Human evaluations on empathetic-dialogues (Rashkin et al., 2018) dataset confirm that MoEL outperforms multitask training baseline in terms of empathy, relevance, and fluency. Furthermore, the case study on generated responses of different Listeners shows high interpretability of our model.

cs.CL↗
arXiv2019

Disentangling spin-orbit coupling and local magnetism in a quasi-two-dimensional electron system

Quantum interference between time-reversed electron paths in two dimensions leads to the well-known weak localization correction to resistance. If spin-orbit coupling is present, the resistance correction is negative, termed weak anti-localization (WAL). Here we report the observation of WAL coexisting with exchange coupling between itinerant electrons and localized magnetic moments. We use low-temperature magneto-transport measurements to investigate the quasi-two-dimensional, high-electron-density interface formed between SrTiO$_3$ (STO) and the anti-ferromagnetic Mott insulator NdTiO$_3$ (NTO). As the magnetic field angle is gradually tilted away from the sample normal, the data reveals the interplay between strong $k$-cubic Rashba-type spin-orbit coupling and a substantial magnetic exchange interaction from local magnetic regions. The resulting quantum corrections to the conduction are in excellent agreement with existing models and allow sensitive determination of the small magnetic moments (22 $μ_B$ on average), their magnetic anisotropy and mutual coupling strength. This effect is expected to arise in other 2D magnetic materials systems.

cond-mat.mes-hall↗
arXiv2019

Getting To Know You: User Attribute Extraction from Dialogues

User attributes provide rich and useful information for user understanding, yet structured and easy-to-use attributes are often sparsely populated. In this paper, we leverage dialogues with conversational agents, which contain strong suggestions of user information, to automatically extract user attributes. Since no existing dataset is available for this purpose, we apply distant supervision to train our proposed two-stage attribute extractor, which surpasses several retrieval and generation baselines on human evaluation. Meanwhile, we discuss potential applications (e.g., personalized recommendation and dialogue systems) of such extracted user attributes, and point out current limitations to cast light on future work.

cs.CL↗
arXiv2019

Multi-view Clustering with the Cooperation of Visible and Hidden Views

Multi-view data are becoming common in real-world modeling tasks and many multi-view data clustering algorithms have thus been proposed. The existing algorithms usually focus on the cooperation of different views in the original space but neglect the influence of the hidden information among these different visible views, or they only consider the hidden information between the views. The algorithms are therefore not efficient since the available information is not fully excavated, particularly the otherness information in different views and the consistency information between them. In practice, the otherness and consistency information in multi-view data are both very useful for effective clustering analyses. In this study, a Multi-View clustering algorithm developed with the Cooperation of Visible and Hidden views, i.e., MV-Co-VH, is proposed. The MV-Co-VH algorithm first projects the multiple views from different visible spaces to the common hidden space by using the non-negative matrix factorization (NMF) strategy to obtain the common hidden view data. Collaborative learning is then implemented in the clustering procedure based on the visible views and the shared hidden view. The results of extensive experiments on UCI multi-view datasets and real-world image multi-view datasets show that the clustering performance of the proposed algorithm is competitive with or even better than that of the existing algorithms.

cs.LG↗
arXiv2019

Multi-View Fuzzy Clustering with The Alternative Learning between Shared Hidden Space and Partition

As the multi-view data grows in the real world, multi-view clus-tering has become a prominent technique in data mining, pattern recognition, and machine learning. How to exploit the relation-ship between different views effectively using the characteristic of multi-view data has become a crucial challenge. Aiming at this, a hidden space sharing multi-view fuzzy clustering (HSS-MVFC) method is proposed in the present study. This method is based on the classical fuzzy c-means clustering model, and obtains associ-ated information between different views by introducing shared hidden space. Especially, the shared hidden space and the fuzzy partition can be learned alternatively and contribute to each other. Meanwhile, the proposed method uses maximum entropy strategy to control the weights of different views while learning the shared hidden space. The experimental result shows that the proposed multi-view clustering method has better performance than many related clustering methods.

cs.LG↗
arXiv2019

Efficient generation of many-body entangled states by multilevel oscillations

We generate high-fidelity massively entangled states in an antiferromagnetic spin-1 Bose-Einstein condensate (BEC) by utilizing multilevel oscillations. Combining the multilevel oscillations with additional adiabatic drives, we greatly shorten the necessary evolution time and relax the requirement on the control accuracy of quadratic Zeeman splitting, from micro-Gauss to milli-Gauss, for a $^{23}$Na spinor BEC. The achieved high fidelities over $96\%$ show that two kinds of massively entangled states, the many-body singlet state and the twin-Fock state, are almost perfectly generated. The generalized spin squeezing parameter drops to a value far below the standard quantum limit even with the presence of atom number fluctuations and stray magnetic fields, illustrating the robustness of our protocol under real experimental conditions. The generated many-body entangled states can be employed to achieve the Heisenberg-limit quantum precision measurement and to attack nonclassical problems in quantum information science.

quant-ph↗
arXiv2019

Revealing Energy Dependence of Quantum Defects via Two Heteronuclear Atoms in an Optical Tweezer

As a physically motivated and computationally simple model for cold atomic and molecular collisions, the multichannel quantum defect theory (MQDT) with frame transformation (FT) formalism provides an analytical treatment of scattering resonances in an arbitrary partial wave between alkali-metal atoms, leading to the experimental observation of $p-$ and $d-$wave resonances. However, the inconsistency of quantum defects for describing scattering resonances shows up when compared with experiments. Here, with two heteronuclear atoms in the ground state of an optical tweezer, the energy dependence of quantum defects is obviously revealed by comparing the measured s-wave scattering length with the prediction of MQDT-FT. By dividing the quantum defects into energy sensitive and insensitive categories, the inconsistency is ultimately removed while retaining the analytic structure of MQDT-FT. This study represents a significant improvement in the analytical MQDT-FT and demonstrates that a clean two-particle system is valuable to the test of collisional physics.

quant-ph↗
arXiv2019

Balanced Coherence Times of Mixed-Species Atomic Qubits in a Dual $3\times3$ Magic-Intensity Optical Dipole Trap Array

In this work, we construct a polarization-mediated magic-intensity (MI) optical dipole trap (ODT) array, in which the detrimental effects of light shifts on the mixed-species qubits are efficiently mitigated so that the coherence times of the mixed-species qubits are both substantially enhanced and balanced for the first time. This mixed-species magic trapping technique relies on the tunability of the coefficient of the third-order cross term and ground state hyperpolarizability, which are inherently dependent on the degree of circular polarization of the trap laser. Experimentally, polarization of the ODT array for $^{85}$Rb qubits is finely adjusted to a definite value so that its working magnetic field required for magic trapping amounts to the one required for magically trapping $^{87}$Rb qubits in another ODT array with fully circular polarization. Ultimately, in such a polarization-mediated MI-ODT array, the coherence times of $^{87}$Rb and $^{85}$Rb qubits are respectively enhanced up to 891$\pm$47 ms and 943$\pm$35 ms. Furthermore, a new source of dephasing effect is revealed, which arises from the noise of the elliptic polarization, and the reduction in corresponding dephasing effect on the $^{85}$Rb qubits is attainable by use of shallow magic intensity. It is anticipated that the novel mixed-species MI-ODT array is a versatile platform for building scalable quantum computers with neutral atoms.

quant-ph↗
arXiv2019

A Cross-Domain Transferable Neural Coherence Model

Coherence is an important aspect of text quality and is crucial for ensuring its readability. One important limitation of existing coherence models is that training on one domain does not easily generalize to unseen categories of text. Previous work advocates for generative models for cross-domain generalization, because for discriminative models, the space of incoherent sentence orderings to discriminate against during training is prohibitively large. In this work, we propose a local discriminative neural model with a much smaller negative sampling space that can efficiently learn against incorrect orderings. The proposed coherence model is simple in structure, yet it significantly outperforms previous state-of-art methods on a standard benchmark dataset on the Wall Street Journal corpus, as well as in multiple new challenging settings of transfer to unseen categories of discourse on Wikipedia articles.

cs.CL↗
arXiv2019

orthoDr: Semiparametric Dimension Reduction via Orthogonality Constrained Optimization

orthoDr is a package in R that solves dimension reduction problems using orthogonality constrained optimization approach. The package serves as a unified framework for many regression and survival analysis dimension reduction models that utilize semiparametric estimating equations. The main computational machinery of orthoDr is a first-order algorithm developed by \cite{wen2013feasible} for optimization within the Stiefel manifold. We implement the algorithm through Rcpp and OpenMP for fast computation. In addition, we developed a general-purpose solver for such constrained problems with user-specified objective functions, which works as a drop-in version of optim(). The package also serves as a platform for future methodology developments along this line of work.

stat.CO↗
arXiv2019

CAiRE_HKUST at SemEval-2019 Task 3: Hierarchical Attention for Dialogue Emotion Classification

Detecting emotion from dialogue is a challenge that has not yet been extensively surveyed. One could consider the emotion of each dialogue turn to be independent, but in this paper, we introduce a hierarchical approach to classify emotion, hypothesizing that the current emotional state depends on previous latent emotions. We benchmark several feature-based classifiers using pre-trained word and emotion embeddings, state-of-the-art end-to-end neural network models, and Gaussian processes for automatic hyper-parameter search. In our experiments, hierarchical architectures consistently give significant improvements, and our best model achieves a 76.77% F1-score on the test set.

cs.CL↗
arXiv2019

Passage Ranking with Weak Supervision

In this paper, we propose a \textit{weak supervision} framework for neural ranking tasks based on the data programming paradigm \citep{Ratner2016}, which enables us to leverage multiple weak supervision signals from different sources. Empirically, we consider two sources of weak supervision signals, unsupervised ranking functions and semantic feature similarities. We train a BERT-based passage-ranking model (which achieves new state-of-the-art performances on two benchmark datasets with full supervision) in our weak supervision framework. Without using ground-truth training labels, BERT-PR models outperform BM25 baseline by a large margin on all three datasets and even beat the previous state-of-the-art results with full supervision on two of the datasets.

cs.IR↗
arXiv2019

ZAIGA: Zhaoshan Long-baseline Atom Interferometer Gravitation Antenna

The Zhaoshan long-baseline Atom Interferometer Gravitation Antenna (ZAIGA) is a new type of underground laser-linked interferometer facility, and is currently under construction. It is in the 200-meter-on-average underground of a mountain named Zhaoshan which is about 80 km southeast to Wuhan. ZAIGA will be equipped with long-baseline atom interferometers, high-precision atom clocks, and large-scale gyros. ZAIGA facility will take an equilateral triangle configuration with two 1-km-apart atom interferometers in each arm, a 300-meter vertical tunnel with atom fountain and atom clocks mounted, and a tracking-and-ranging 1-km-arm-length prototype with lattice optical clocks linked by locked lasers. The ZAIGA facility will be used for experimental research on gravitation and related problems including gravitational wave detection, high-precision test of the equivalence principle of micro-particles, clock based gravitational red-shift measurement, rotation measurement and gravito-magnetic effect.

physics.atom-ph↗
arXiv2019

Multi-view Information-theoretic Co-clustering for Co-occurrence Data

Multi-view clustering has received much attention recently. Most of the existing multi-view clustering methods only focus on one-sided clustering. As the co-occurring data elements involve the counts of sample-feature co-occurrences, it is more efficient to conduct two-sided clustering along the samples and features simultaneously. To take advantage of two-sided clustering for the co-occurrences in the scene of multi-view clustering, a two-sided multi-view clustering method is proposed, i.e., multi-view information-theoretic co-clustering (MV-ITCC). The proposed method realizes two-sided clustering for co-occurring multi-view data under the formulation of information theory. More specifically, it exploits the agreement and disagreement among views by sharing a common clustering results along the sample dimension and keeping the clustering results of each view specific along the feature dimension. In addition, the mechanism of maximum entropy is also adopted to control the importance of different views, which can give a right balance in leveraging the agreement and disagreement. Extensive experiments are conducted on text and image multi-view datasets. The results clearly demonstrate the superiority of the proposed method.

cs.LG↗
arXiv2019

Kaleido: An Efficient Out-of-core Graph Mining System on A Single Machine

Graph mining is one of the most important categories of graph algorithms. However, exploring the subgraphs of an input graph produces a huge amount of intermediate data. The 'think like a vertex' programming paradigm, pioneered by Pregel, cannot readily formulate mining problems, which is designed to produce graph computation problems like PageRank. Existing mining systems like Arabesque and RStream need large amounts of computing and memory resources. In this paper, we present Kaleido, an efficient single machine, out-of-core graph mining system which treats disks as an extension of memory. Kaleido treats intermediate data in graph mining tasks as a tensor and adopts a succinct data structure for the intermediate data. Kaleido utilizes the eigenvalue of the adjacency matrix of a subgraph to efficiently solve the subgraph isomorphism problems with an acceptable constraint that the vertex number of a subgraph is less than 9. Kaleido implements half-memory-half-disk storage for storing large intermediate data, which treats the disk as an extension of the memory. Comparing with two state-of-the-art mining systems, Arabesque and RStream, Kaleido outperforms them by a GeoMean 12.3$\times$ and 40.0$\times$ respectively.

cs.DC↗
arXiv2019

Joint Information Preservation for Heterogeneous Domain Adaptation

Domain adaptation aims to assist the modeling tasks of the target domain with knowledge of the source domain. The two domains often lie in different feature spaces due to diverse data collection methods, which leads to the more challenging task of heterogeneous domain adaptation (HDA). A core issue of HDA is how to preserve the information of the original data during adaptation. In this paper, we propose a joint information preservation method to deal with the problem. The method preserves the information of the original data from two aspects. On the one hand, although paired samples often exist between the two domains of the HDA, current algorithms do not utilize such information sufficiently. The proposed method preserves the paired information by maximizing the correlation of the paired samples in the shared subspace. On the other hand, the proposed method improves the strategy of preserving the structural information of the original data, where the local and global structural information are preserved simultaneously. Finally, the joint information preservation is integrated by distribution matching. Experimental results show the superiority of the proposed method over the state-of-the-art HDA algorithms.

cs.LG↗