arXiv subjects

Yiqun Liu

Publications and source records attributed to Yiqun Liu.

At least 181 records · Page 10Linked to original sources

arXiv2021

SelfGait: A Spatiotemporal Representation Learning Method for Self-supervised Gait Recognition

Gait recognition plays a vital role in human identification since gait is a unique biometric feature that can be perceived at a distance. Although existing gait recognition methods can learn gait features from gait sequences in different ways, the performance of gait recognition suffers from insufficient labeled data, especially in some practical scenarios associated with short gait sequences or various clothing styles. It is unpractical to label the numerous gait data. In this work, we propose a self-supervised gait recognition method, termed SelfGait, which takes advantage of the massive, diverse, unlabeled gait data as a pre-training process to improve the representation abilities of spatiotemporal backbones. Specifically, we employ the horizontal pyramid mapping (HPM) and micro-motion template builder (MTB) as our spatiotemporal backbones to capture the multi-scale spatiotemporal representations. Experiments on CASIA-B and OU-MVLP benchmark gait datasets demonstrate the effectiveness of the proposed SelfGait compared with four state-of-the-art gait recognition methods. The source code has been released at https://github.com/EchoItLiu/SelfGait.

cs.CV↗

arXiv2021

Anisotropic dressing of electrons in electron-doped cuprate superconductors

The recent experiments revealed a remarkable possibility for the absence of the disparity between the phase diagrams of the electron- and hole-doped cuprate superconductors, while such an aspect should be also reflected in the dressing of the electrons. Here the phase diagram of the electron-doped cuprate superconductors and the related exotic features of the anisotropic dressing of the electrons are studied based on the kinetic-energy driven superconductivity. It is shown that although the optimized Tc in the electron-doped side is much smaller than that in the hole-doped case, the electron- and hole-doped cuprate superconductors rather resemble each other in the doping range of the superconducting dome, indicating an absence of the disparity between the phase diagrams of the electron- and hole-doped cuprate superconductors. In particular, the anisotropic dressing of the electrons due to the strong electron's coupling to a strongly dispersive spin excitation leads to that the electron Fermi surface is truncated to form the disconnected Fermi arcs centered around the nodal region. Concomitantly, the dip in the peak-dip-hump structure of the quasiparticle excitation spectrum is directly associated with the corresponding peak in the quasiparticle scattering rate, while the dispersion kink is always accompanied by the corresponding inflection point in the total self-energy, as the dip in the peak-dip-hump structure and dispersion kink in the hole-doped counterparts. The theory also predicts that both the normal and anomalous self-energies exhibit the well-pronounced low-energy peak-structures.

cond-mat.supr-con↗

arXiv2021

Peak-structure in self-energy of cuprate superconductors

The recently deduced normal and anomalous self-energies from photoemission spectra of cuprate superconductors via the machine learning technique are calling for an explanation. Here the normal and anomalous self-energies in cuprate superconductors are analyzed within the framework of the kinetic-energy-driven superconductivity. It is shown that the exchanged spin excitations give rise to the well-pronounced low-energy peak-structures in both the normal and anomalous self-energies, however, they do not cancel in the total self-energy. In particular, the peak-structure in the normal self-energy is mainly responsible for the peak-dip-hump structure in the single-particle excitation spectrum, and can persist into the normal-state, while the sharp peak in the anomalous self-energy gives rise to a crucial contribution to the superconducting gap, and vanishes in the normal-state. Moreover, the evolution of the peak-structure with doping and momentum are also analyzed.

cond-mat.supr-con↗

arXiv2020

THUIR@COLIEE-2020: Leveraging Semantic Understanding and Exact Matching for Legal Case Retrieval and Entailment

In this paper, we present our methodologies for tackling the challenges of legal case retrieval and entailment in the Competition on Legal Information Extraction / Entailment 2020 (COLIEE-2020). We participated in the two case law tasks, i.e., the legal case retrieval task and the legal case entailment task. Task 1 (the retrieval task) aims to automatically identify supporting cases from the case law corpus given a new case, and Task 2 (the entailment task) to identify specific paragraphs that entail the decision of a new case in a relevant case. In both tasks, we employed the neural models for semantic understanding and the traditional retrieval models for exact matching. As a result, our team (TLIR) ranked 2nd among all of the teams in Task 1 and 3rd among teams in Task 2. Experimental results suggest that combing models of semantic understanding and exact matching benefits the legal case retrieval task while the legal case entailment task relies more on semantic understanding.

cs.IR↗

arXiv2020

Learning To Retrieve: How to Train a Dense Retrieval Model Effectively and Efficiently

Ranking has always been one of the top concerns in information retrieval research. For decades, lexical matching signal has dominated the ad-hoc retrieval process, but it also has inherent defects, such as the vocabulary mismatch problem. Recently, Dense Retrieval (DR) technique has been proposed to alleviate these limitations by capturing the deep semantic relationship between queries and documents. The training of most existing Dense Retrieval models relies on sampling negative instances from the corpus to optimize a pairwise loss function. Through investigation, we find that this kind of training strategy is biased and fails to optimize full retrieval performance effectively and efficiently. To solve this problem, we propose a Learning To Retrieve (LTRe) training technique. LTRe constructs the document index beforehand. At each training iteration, it performs full retrieval without negative sampling and then updates the query representation model parameters. Through this process, it teaches the DR model how to retrieve relevant documents from the entire corpus instead of how to rerank a potentially biased sample of documents. Experiments in both passage retrieval and document retrieval tasks show that: 1) in terms of effectiveness, LTRe significantly outperforms all competitive sparse and dense baselines. It even gains better performance than the BM25-BERT cascade system under reasonable latency constraints. 2) in terms of training efficiency, compared with the previous state-of-the-art DR method, LTRe provides more than 170x speed-up in the training process. Training with a compressed index further saves computing resources with minor performance loss.

cs.IR↗

arXiv2020

Practical Deep Raw Image Denoising on Mobile Devices

Deep learning-based image denoising approaches have been extensively studied in recent years, prevailing in many public benchmark datasets. However, the stat-of-the-art networks are computationally too expensive to be directly applied on mobile devices. In this work, we propose a light-weight, efficient neural network-based raw image denoiser that runs smoothly on mainstream mobile devices, and produces high quality denoising results. Our key insights are twofold: (1) by measuring and estimating sensor noise level, a smaller network trained on synthetic sensor-specific data can out-perform larger ones trained on general data; (2) the large noise level variation under different ISO settings can be removed by a novel k-Sigma Transform, allowing a small network to efficiently handle a wide range of noise levels. We conduct extensive experiments to demonstrate the efficiency and accuracy of our approach. Our proposed mobile-friendly denoising model runs at ~70 milliseconds per megapixel on Qualcomm Snapdragon 855 chipset, and it is the basis of the night shot feature of several flagship smartphones released in 2019.

eess.IV↗

arXiv2020

An Empirical Study of Clarifying Question-Based Systems

Search and recommender systems that take the initiative to ask clarifying questions to better understand users' information needs are receiving increasing attention from the research community. However, to the best of our knowledge, there is no empirical study to quantify whether and to what extent users are willing or able to answer these questions. In this work, we conduct an online experiment by deploying an experimental system, which interacts with users by asking clarifying questions against a product repository. We collect both implicit interaction behavior data and explicit feedback from users showing that: (a) users are willing to answer a good number of clarifying questions (11-21 on average), but not many more than that; (b) most users answer questions until they reach the target product, but also a fraction of them stops due to fatigue or due to receiving irrelevant questions; (c) part of the users' answers (12-17%) are actually opposite to the description of the target product; while (d) most of the users (66-84%) find the question-based system helpful towards completing their tasks. Some of the findings of the study contradict current assumptions on simulated evaluations in the field, while they point towards improvements in the evaluation framework and can inspire future interactive search/recommender system designs.

cs.IR↗

arXiv2020

RepBERT: Contextualized Text Embeddings for First-Stage Retrieval

Although exact term match between queries and documents is the dominant method to perform first-stage retrieval, we propose a different approach, called RepBERT, to represent documents and queries with fixed-length contextualized embeddings. The inner products of query and document embeddings are regarded as relevance scores. On MS MARCO Passage Ranking task, RepBERT achieves state-of-the-art results among all initial retrieval techniques. And its efficiency is comparable to bag-of-words methods.

cs.IR↗

arXiv2020

STAS: Adaptive Selecting Spatio-Temporal Deep Features for Improving Bias Correction on Precipitation

Numerical Weather Prediction (NWP) can reduce human suffering by predicting disastrous precipitation in time. A commonly-used NWP in the world is the European Centre for medium-range weather forecasts (EC). However, it is necessary to correct EC forecast through Bias Correcting on Precipitation (BCoP) since we still have not fully understood the mechanism of precipitation, making EC often have some biases. The existing BCoPs suffers from limited prior data and the fixed Spatio-Temporal (ST) scale. We thus propose an end-to-end deep-learning BCoP model named Spatio-Temporal feature Auto-Selective (STAS) model to select optimal ST regularity from EC via the ST Feature-selective Mechanisms (SFM/TFM). Given different input features, these two mechanisms can automatically adjust the spatial and temporal scales for correcting. Experiments on an EC public dataset indicate that compared with 8 published BCoP methods, STAS shows state-of-the-art performance on several criteria of BCoP, named threat scores (TS). Further, ablation studies justify that the SFM/TFM indeed work well in boosting the performance of BCoP, especially on the heavy precipitation.

cs.LG↗

arXiv2020

Renormalization of electrons in bilayer cuprate superconductors

The characteristic features of the renormalization of the electrons in the bilayer cuprate superconductors are investigated within the kinetic-energy driven superconductivity. It is shown that the quasiparticle excitation spectrum is split into its bonding and antibonding components due to the presence of the bilayer coupling, with each component that is independent. However, in the underdoped and optimally doped regimes, although the bonding and antibonding electron Fermi surface (EFS) contours deriving from the bonding and antibonding layers are truncated to form the bonding and antibonding Fermi arcs, almost all spectral weights in the bonding and antibonding Fermi arcs are reduced to the tips of the bonding and antibonding Fermi arcs, which in this case coincide with the bonding and antibonding hot spots. These hot spots connected by the scattering wave vectors ${\bf q}_{i} $ construct an octet scattering model, and then the enhancement of the quasiparticle scattering processes with the scattering wave vectors ${\bf q}_{i}$ is confirmed via the result of the autocorrelation of the ARPES spectral intensities. Moreover, the peak-dip-hump (PDH) structure developed in each component of the quasiparticle excitation spectrum along the corresponding EFS is directly related with the peak structure in the quasiparticle scattering rate except for at around the hot spots, where the PDH structure is caused mainly by the bilayer coupling. Although the kink in the quasiparticle dispersion is present all around EFS, when the momentum moves away from the node to the antinode, the kink energy smoothly decreases, while the dispersion kink becomes more pronounced, and in particular, near the cut close to the antinode, develops into a break separating of the fasting dispersing high-energy part of the quasiparticle excitation spectrum from the slower dispersing low-energy part.

cond-mat.supr-con↗

arXiv2020

ARPES autocorrelation in electron-doped cuprate superconductors

The angle-resolved photoemission spectroscopy (ARPES) autocorrelation in the electron-doped cuprate superconductors is studied based on the kinetic-energy driven superconducting (SC) mechanism. It is shown that the strong electron correlation induces the electron Fermi surface (EFS) reconstruction, where the most of the quasiparticles locate at around the hot spots on EFS, and then these hot spots connected by the scattering wave vectors ${\bf q}_{i}$ construct an {\it octet} scattering model. In a striking analogy to the hole-doped case, the sharp ARPES autocorrelation peaks are directly correlated with the scattering wave vectors ${\bf q}_{i}$, and are weakly dispersive in momentum space. However, in a clear contrast to the hole-doped counterparts, the position of the ARPES autocorrelation peaks move toward to the opposite direction with the increase of doping. The theory also indicates that there is an intrinsic connection between the ARPES autocorrelation and quasiparticle scattering interference (QSI) in the electron-doped cuprate superconductors.

cond-mat.supr-con↗

arXiv2019

Towards a Precipitation Bias Corrector against Noise and Maldistribution

With broad applications in various public services like aviation management and urban disaster warning, numerical precipitation prediction plays a crucial role in weather forecast. However, constrained by the limitation of observation and conventional meteorological models, the numerical precipitation predictions are often highly biased. To correct this bias, classical correction methods heavily depend on profound experts who have knowledge in aerodynamics, thermodynamics and meteorology. As precipitation can be influenced by countless factors, however, the performances of these expert-driven methods can drop drastically when some un-modeled factors change. To address this issue, this paper presents a data-driven deep learning model which mainly includes two blocks, i.e. a Denoising Autoencoder Block and an Ordinal Regression Block. To the best of our knowledge, it is the first expert-free models for bias correction. The proposed model can effectively correct the numerical precipitation prediction based on 37 basic meteorological data from European Centre for Medium-Range Weather Forecasts (ECMWF). Experiments indicate that compared with several classical machine learning algorithms and deep learning models, our method achieves the best correcting performance and meteorological index, namely the threat scores (TS), obtaining satisfactory visualization effect.

cs.LG↗

arXiv2019

Doping dependence of electromagnetic response in cuprate superconductors

The study of the electromagnetic response in cuprate superconductors plays a crucial role in the understanding of the essential physics of these materials. Here the doping dependence of the electromagnetic response in cuprate superconductors is studied within the kinetic-energy driven superconducting mechanism. The kernel of the response function is evaluated based on the linear response approximation for a purely transverse vector potential, and can be broken up into its diamagnetic and paramagnetic parts. In particular, this paramagnetic part exactly cancels the corresponding diamagnetic part in the normal-state, and then the Meissner effect is obtained within the entire superconducting phase. Following this kernel of the response function, the electromagnetic response calculation in terms of the specular reflection model qualitatively reproduces many of the striking features observed in the experiments. In particular, the local magnetic-field profile follows an exponential law, while the superfluid density exhibits the nonlinear temperature behavior at the lowest temperatures, followed by the linear temperature dependence extending over the most of the superconducting temperature range. Moreover, the maximal value of the superfluid density occurs at around the critical doping $δ_{\rm critical}\sim 0.16$, and then decreases in both lower doped and higher doped regimes. The theory also shows that the nonlinear temperature dependence of the superfluid density at the lowest temperatures can be attributed to the nonlocal effects induced by the d-wave gap nodes on the electron Fermi surface.

cond-mat.supr-con↗

arXiv2019

Doping and momentum dependence of coupling strength in cuprate superconductors

Superconductivity is caused by the interaction between electrons by the exchange of collective bosonic excitations, however, this bosonic glue forming electron pairs is manifested itself by the coupling strength of the electrons to collective bosonic excitations. Here the doping and momentum dependence of the coupling strength of the electrons to spin excitations in cuprate superconductors is studied within the framework of the kinetic-energy-driven superconducting mechanism. The normal self-energy in the particle-hole channel and pairing self-energy in the particle-pariticle channel generated by the interaction between electrons by the exchange of spin excitation are employed to extract the coupling strengths of the electrons to spin excitations in the particle-hole and particle-particle channels, respectively. It is shown that below the superconducting transition temperature, both the coupling strengths in the particle-hole and particle-particle channels around the antinodes consist of two peaks, with a sharp low-energy peak located at around 5 meV in the optimally doped regime, and a broad band with a weak peak centered at around 40 meV. In particular, this two-peak structure in the coupling strength in the particle-hole channel can persist into the normal-state, while as a consequence of the d-wave type symmetry of the superconducting gap, the coupling strength in the particle-particle channel vanishes at the nodes. However, the positions of the peaks in the coupling strengths in the underdoped regime shift towards to higher energies with the increase of doping. More specifically, although the positions of the peaks in the coupling strengths move to lower energies from the antinode to the hot spot on the electron Fermi surface, the weights of the peaks decrease smoothly with the move of the momentum from the antinode to the hot spot, and fade away at the hot spots.

cond-mat.supr-con↗

arXiv2019

Jointly Learning Explainable Rules for Recommendation with Knowledge Graph

Explainability and effectiveness are two key aspects for building recommender systems. Prior efforts mostly focus on incorporating side information to achieve better recommendation performance. However, these methods have some weaknesses: (1) prediction of neural network-based embedding methods are hard to explain and debug; (2) symbolic, graph-based approaches (e.g., meta path-based models) require manual efforts and domain knowledge to define patterns and rules, and ignore the item association types (e.g. substitutable and complementary). In this paper, we propose a novel joint learning framework to integrate \textit{induction of explainable rules from knowledge graph} with \textit{construction of a rule-guided neural recommendation model}. The framework encourages two modules to complement each other in generating effective and explainable recommendation: 1) inductive rules, mined from item-centric knowledge graphs, summarize common multi-hop relational patterns for inferring different item associations and provide human-readable explanation for model prediction; 2) recommendation module can be augmented by induced rules and thus have better generalization ability dealing with the cold-start issue. Extensive experiments\footnote{Code and data can be found at: \url{https://github.com/THUIR/RuleRec}} show that our proposed method has achieved significant improvements in item recommendation over baselines on real-world datasets. Our model demonstrates robust performance over "noisy" item knowledge graphs, generated by linking item names to related entities.

cs.IR↗

arXiv2019

Temporal Relational Ranking for Stock Prediction

Stock prediction aims to predict the future trends of a stock in order to help investors to make good investment decisions. Traditional solutions for stock prediction are based on time-series models. With the recent success of deep neural networks in modeling sequential data, deep learning has become a promising choice for stock prediction. However, most existing deep learning solutions are not optimized towards the target of investment, i.e., selecting the best stock with the highest expected revenue. Specifically, they typically formulate stock prediction as a classification (to predict stock trend) or a regression problem (to predict stock price). More importantly, they largely treat the stocks as independent of each other. The valuable signal in the rich relations between stocks (or companies), such as two stocks are in the same sector and two companies have a supplier-customer relation, is not considered. In this work, we contribute a new deep learning solution, named Relational Stock Ranking (RSR), for stock prediction. Our RSR method advances existing solutions in two major aspects: 1) tailoring the deep learning models for stock ranking, and 2) capturing the stock relations in a time-sensitive manner. The key novelty of our work is the proposal of a new component in neural network modeling, named Temporal Graph Convolution, which jointly models the temporal evolution and relation network of stocks. To validate our method, we perform back-testing on the historical data of two stock markets, NYSE and NASDAQ. Extensive experiments demonstrate the superiority of our RSR method. It outperforms state-of-the-art stock prediction solutions achieving an average return ratio of 98% and 71% on NYSE and NASDAQ, respectively.

cs.CE↗

arXiv2018

Autocorrelation of quasiparticle spectral intensities and its connection with quasiparticle scattering interference in cuprate superconductors

The quasiparticle excitation is one of the most fundamental and ubiquitous physical observables in cuprate superconductors, carrying information about the bosonic glue forming electron pairs. Here the autocorrelation of the quasiparticle excitation spectral intensities in cuprate superconductors and its connection with the quasiparticle scattering interference are investigated based on the framework of the kinetic-energy driven superconducting mechanism by taking into account the pseudogap effect. It is shown that the octet scattering model of the quasiparticle scattering processes with the scattering wave vectors ${\bf q}_{i}$ connecting the hot spots on the constant energy contours is intrinsically related to the emergence of the highly anisotropic momentum-dependence of the pseudogap. Concomitantly, the sharp peaks in the autocorrelation of the quasiparticle excitation spectral intensities with the wave vectors ${\bf q}_{i}$ are directly correlated to the regions of the highest joint density of states. Moreover, the momentum-space structure of the autocorrelation patterns of the quasiparticle excitation spectral intensities is well consistent with the momentum-space structure of the quasiparticle scattering interference patterns observed from Fourier-transform scanning tunneling spectroscopy experiments. The theory therefore confirms an intimate connection between the angle-resolved photoemission spectroscopy autocorrelation and quasiparticle scattering interference in cuprate superconductors.

cond-mat.supr-con↗

arXiv2018

Hidden pair-density-wave order in cuprate superconductors

When the Mott insulating state is suppressed by charge carrier doping, the pseudogap phenomenon emerges, where at the low-temperature limit, superconductivity coexists with some ordered electronic states. Within the framework of the kinetic-energy-driven superconductivity, the nature of the pair-density-wave order in cuprate superconductors is studied by taking into account the pseudogap effect. It is shown that the onset of the pair-density-wave order does not produce an ordered gap, but rather a novel hidden order as a result of the interplay of the charge-density-wave order with superconductivity. As a consequence, this novel hidden pair-density-wave order as a subsidiary order parameter coexists with the charge-density-wave order in the superconducting-state, and is absent from the normal-state.

cond-mat.supr-con↗