arXiv subjects

Qiang Liu

Publications and source records attributed to Qiang Liu.

At least 415 records · Page 23Linked to original sources

arXiv2020

Modeling Dyadic Conversations for Personality Inference

Nowadays, automatical personality inference is drawing extensive attention from both academia and industry. Conventional methods are mainly based on user generated contents, e.g., profiles, likes, and texts of an individual, on social media, which are actually not very reliable. In contrast, dyadic conversations between individuals can not only capture how one expresses oneself, but also reflect how one reacts to different situations. Rich contextual information in dyadic conversation can explain an individual's response during his or her conversation. In this paper, we propose a novel augmented Gated Recurrent Unit (GRU) model for learning unsupervised Personal Conversational Embeddings (PCE) based on dyadic conversations between individuals. We adjust the formulation of each layer of a conventional GRU with sequence to sequence learning and personal information of both sides of the conversation. Based on the learned PCE, we can infer the personality of each individual. We conduct experiments on the Movie Script dataset, which is collected from conversations between characters in movie scripts. We find that modeling dyadic conversations between individuals can significantly improve personality inference accuracy. Experimental results illustrate the successful performance of our proposed method.

cs.CL↗

arXiv2020

Deep Active Learning by Model Interpretability

Recent successes of Deep Neural Networks (DNNs) in a variety of research tasks, however, heavily rely on the large amounts of labeled samples. This may require considerable annotation cost in real-world applications. Fortunately, active learning is a promising methodology to train high-performing model with minimal annotation cost. In the deep learning context, the critical question of active learning is how to precisely identify the informativeness of samples for DNN. In this paper, inspired by piece-wise linear interpretability in DNN, we introduce the linearly separable regions of samples to the problem of active learning, and propose a novel Deep Active learning approach by Model Interpretability (DAMI). To keep the maximal representativeness of the entire unlabeled data, DAMI tries to select and label samples on different linearly separable regions introduced by the piece-wise linear interpretability in DNN. We focus on modeling Multi-Layer Perception (MLP) for modeling tabular data. Specifically, we use the local piece-wise interpretation in MLP as the representation of each sample, and directly run K-Center clustering to select and label samples. To be noted, this whole process of DAMI does not require any hyper-parameters to tune manually. To verify the effectiveness of our approach, extensive experiments have been conducted on several tabular datasets. The experimental results demonstrate that DAMI constantly outperforms several state-of-the-art compared approaches.

cs.LG↗

arXiv2020

DeepSlicing: Deep Reinforcement Learning Assisted Resource Allocation for Network Slicing

Network slicing enables multiple virtual networks run on the same physical infrastructure to support various use cases in 5G and beyond. These use cases, however, have very diverse network resource demands, e.g., communication and computation, and various performance metrics such as latency and throughput. To effectively allocate network resources to slices, we propose DeepSlicing that integrates the alternating direction method of multipliers (ADMM) and deep reinforcement learning (DRL). DeepSlicing decomposes the network slicing problem into a master problem and several slave problems. The master problem is solved based on convex optimization and the slave problem is handled by DRL method which learns the optimal resource allocation policy. The performance of the proposed algorithm is validated through network simulations.

eess.SP↗

arXiv2020

Go Wide, Then Narrow: Efficient Training of Deep Thin Networks

For deploying a deep learning model into production, it needs to be both accurate and compact to meet the latency and memory constraints. This usually results in a network that is deep (to ensure performance) and yet thin (to improve computational efficiency). In this paper, we propose an efficient method to train a deep thin network with a theoretic guarantee. Our method is motivated by model compression. It consists of three stages. First, we sufficiently widen the deep thin network and train it until convergence. Then, we use this well-trained deep wide network to warm up (or initialize) the original deep thin network. This is achieved by layerwise imitation, that is, forcing the thin network to mimic the intermediate outputs of the wide network from layer to layer. Finally, we further fine tune this already well-initialized deep thin network. The theoretical guarantee is established by using the neural mean field analysis. It demonstrates the advantage of our layerwise imitation approach over backpropagation. We also conduct large-scale empirical experiments to validate the proposed method. By training with our method, ResNet50 can outperform ResNet101, and BERT Base can be comparable with BERT Large, when ResNet101 and BERT Large are trained under the standard training procedures as in the literature.

cs.LG↗

arXiv2020

Uniform Decay Estimates for Solutions of a Class of Retarded Integral Inequalities

Some uniform decay estimates are established for solutions of the following type of retarded integral inequalities: $$y(t)\leq E(t,τ)||y_τ||+\int_τ^t K_1(t,s)||y_s||ds+\int_t^\infty K_2(t,s)||y_s||ds+ρ, \hspace{0.5cm} t\geqτ\geq 0.$$ As a simple example of application, the retarded scalar functional differential equation $\dot x=-a(t)x+B(t,x_t)$ is considered, and the global asymptotic stability of the equation is proved under weaker conditions. Another example is the ODE system $\dot x=F_0(t,x)+\sum_{i=1}^m F_i(t,x(t-r_i(t)))$ on $R^n$ with superlinear nonlinearities $F_i$ ($0\leq i\leq m$). The existence of a global pullback attractor of the system is established under appropriate dissipation conditions. The third example for application concerns the study of the dynamics of the functional cocycle system $\frac{du}{dt}+Au=F(θ_tp,u_t)$ in a Banach space $X$ with sublinear nonlinearity. In particular, the existence and uniqueness of a nonautonomous stationary solution $Γ$ is obtained under the hyperbolicity assumption on operator $A$ and some additional hypotheses, and the global asymptotic stability of $Γ$ is also addressed.

math.DS↗

arXiv2020

Accountable Off-Policy Evaluation With Kernel Bellman Statistics

We consider off-policy evaluation (OPE), which evaluates the performance of a new policy from observed data collected from previous experiments, without requiring the execution of the new policy. This finds important applications in areas with high execution cost or safety concerns, such as medical diagnosis, recommendation systems and robotics. In practice, due to the limited information from off-policy data, it is highly desirable to construct rigorous confidence intervals, not just point estimation, for the policy performance. In this work, we propose a new variational framework which reduces the problem of calculating tight confidence bounds in OPE into an optimization problem on a feasible set that catches the true state-action value function with high probability. The feasible set is constructed by leveraging statistical properties of a recently proposed kernel Bellman loss (Feng et al., 2019). We design an efficient computational approach for calculating our bounds, and extend it to perform post-hoc diagnosis and correction for existing estimators. Empirical results show that our method yields tight confidence intervals in different settings.

cs.LG↗

arXiv2020

Edgeworth corrections for spot volatility estimator

We develop Edgeworth expansion theory for spot volatility estimator under general assumptions on the log-price process that allow for drift and leverage effect. The result is based on further estimation of skewness and kurtosis, when compared with existing second order asymptotic normality result. Thus our theory can provide with a refinement result for the finite sample distribution of spot volatility. We also construct feasible confidence intervals (one-sided and two-sided) for spot volatility by using Edgeworth expansion. The Monte Carlo simulation study we conduct shows that the intervals based on Edgeworth expansion perform better than the conventional intervals based on normal approximation, which justifies the correctness of our theoretical conclusion.

math.ST↗

arXiv2020

Deep Graph Contrastive Representation Learning

Graph representation learning nowadays becomes fundamental in analyzing graph-structured data. Inspired by recent success of contrastive methods, in this paper, we propose a novel framework for unsupervised graph representation learning by leveraging a contrastive objective at the node level. Specifically, we generate two graph views by corruption and learn node representations by maximizing the agreement of node representations in these two views. To provide diverse node contexts for the contrastive objective, we propose a hybrid scheme for generating graph views on both structure and attribute levels. Besides, we provide theoretical justification behind our motivation from two perspectives, mutual information and the classical triplet loss. We perform empirical experiments on both transductive and inductive learning tasks using a variety of real-world datasets. Experimental experiments demonstrate that despite its simplicity, our proposed method consistently outperforms existing state-of-the-art methods by large margins. Moreover, our unsupervised method even surpasses its supervised counterparts on transductive tasks, demonstrating its great potential in real-world applications.

cs.LG↗

arXiv2020

Energy-Aware Neural Architecture Optimization with Fast Splitting Steepest Descent

Designing energy-efficient networks is of critical importance for enabling state-of-the-art deep learning in mobile and edge settings where the computation and energy budgets are highly limited. Recently, Liu et al. (2019) framed the search of efficient neural architectures into a continuous splitting process: it iteratively splits existing neurons into multiple off-springs to achieve progressive loss minimization, thus finding novel architectures by gradually growing the neural network. However, this method was not specifically tailored for designing energy-efficient networks, and is computationally expensive on large-scale benchmarks. In this work, we substantially improve Liu et al. (2019) in two significant ways: 1) we incorporate the energy cost of splitting different neurons to better guide the splitting process, thereby discovering more energy-efficient network architectures; 2) we substantially speed up the splitting process of Liu et al. (2019), which requires expensive eigen-decomposition, by proposing a highly scalable Rayleigh-quotient stochastic gradient algorithm. Our fast algorithm allows us to reduce the computational cost of splitting to the same level of typical back-propagation updates and enables efficient implementation on GPU. Extensive empirical results show that our method can train highly accurate and energy-efficient networks on challenging datasets such as ImageNet, improving a variety of baselines, including the pruning-based methods and expert-designed architectures.

cs.LG↗

arXiv2020

TFNet: Multi-Semantic Feature Interaction for CTR Prediction

The CTR (Click-Through Rate) prediction plays a central role in the domain of computational advertising and recommender systems. There exists several kinds of methods proposed in this field, such as Logistic Regression (LR), Factorization Machines (FM) and deep learning based methods like Wide&Deep, Neural Factorization Machines (NFM) and DeepFM. However, such approaches generally use the vector-product of each pair of features, which have ignored the different semantic spaces of the feature interactions. In this paper, we propose a novel Tensor-based Feature interaction Network (TFNet) model, which introduces an operating tensor to elaborate feature interactions via multi-slice matrices in multiple semantic spaces. Extensive offline and online experiments show that TFNet: 1) outperforms the competitive compared methods on the typical Criteo and Avazu datasets; 2) achieves large improvement of revenue and click rate in online A/B tests in the largest Chinese App recommender system, Tencent MyApp.

cs.IR↗

arXiv2020

A Multi-View Approach Based on Naming Behavioral Modeling for Aligning Chinese User Accounts across Multiple Networks

Hundreds of millions of Chinese people have become social network users in recent years, and aligning the accounts of common Chinese users across multiple social networks is valuable to many inter-network applications, e.g., cross-network recommendation, cross-network link prediction. Many methods have explored the proper ways of utilizing account name information into aligning the common English users' accounts. However, how to properly utilize the account name information when aligning the Chinese user accounts remains to be detailedly studied. In this paper, we firstly discuss the available naming behavioral models as well as the related features for different types of Chinese account name matchings. Secondly, we propose the framework of Multi-View Cross-Network User Alignment (MCUA) method, which uses a multi-view framework to creatively integrate different models to deal with different types of Chinese account name matchings, and can consider all of the studied features when aligning the Chinese user accounts. Finally, we conduct experiments to prove that MCUA can outperform many existing methods on aligning Chinese user accounts between Sina Weibo and Twitter. Besides, we also study the best learning models and the top-k valuable features of different types of name matchings for MCUA over our experimental data sets.

cs.SI↗

arXiv2020

SAFER: A Structure-free Approach for Certified Robustness to Adversarial Word Substitutions

State-of-the-art NLP models can often be fooled by human-unaware transformations such as synonymous word substitution. For security reasons, it is of critical importance to develop models with certified robustness that can provably guarantee that the prediction is can not be altered by any possible synonymous word substitution. In this work, we propose a certified robust method based on a new randomized smoothing technique, which constructs a stochastic ensemble by applying random word substitutions on the input sentences, and leverage the statistical properties of the ensemble to provably certify the robustness. Our method is simple and structure-free in that it only requires the black-box queries of the model outputs, and hence can be applied to any pre-trained models (such as BERT) and any types of models (world-level or subword-level). Our method significantly outperforms recent state-of-the-art methods for certified robustness on both IMDB and Amazon text classification tasks. To the best of our knowledge, we are the first work to achieve certified robustness on large systems such as BERT with practically meaningful certified accuracy.

cs.LG↗

arXiv2020

TAGNN: Target Attentive Graph Neural Networks for Session-based Recommendation

Session-based recommendation nowadays plays a vital role in many websites, which aims to predict users' actions based on anonymous sessions. There have emerged many studies that model a session as a sequence or a graph via investigating temporal transitions of items in a session. However, these methods compress a session into one fixed representation vector without considering the target items to be predicted. The fixed vector will restrict the representation ability of the recommender model, considering the diversity of target items and users' interests. In this paper, we propose a novel target attentive graph neural network (TAGNN) model for session-based recommendation. In TAGNN, target-aware attention adaptively activates different user interests with respect to varied target items. The learned interest representation vector varies with different target items, greatly improving the expressiveness of the model. Moreover, TAGNN harnesses the power of graph neural networks to capture rich item transitions in sessions. Comprehensive experiments conducted on real-world datasets demonstrate its superiority over state-of-the-art methods.

cs.IR↗

arXiv2020

An Empirical Study on Feature Discretization

When dealing with continuous numeric features, we usually adopt feature discretization. In this work, to find the best way to conduct feature discretization, we present some theoretical analysis, in which we focus on analyzing correctness and robustness of feature discretization. Then, we propose a novel discretization method called Local Linear Encoding (LLE). Experiments on two numeric datasets show that, LLE can outperform conventional discretization method with much fewer model parameters.

cs.LG↗

arXiv2020

Regularization Matters: Generalization and Optimization of Neural Nets v.s. their Induced Kernel

Recent works have shown that on sufficiently over-parametrized neural nets, gradient descent with relatively large initialization optimizes a prediction function in the RKHS of the Neural Tangent Kernel (NTK). This analysis leads to global convergence results but does not work when there is a standard $\ell_2$ regularizer, which is useful to have in practice. We show that sample efficiency can indeed depend on the presence of the regularizer: we construct a simple distribution in d dimensions which the optimal regularized neural net learns with $O(d)$ samples but the NTK requires $Ω(d^2)$ samples to learn. To prove this, we establish two analysis tools: i) for multi-layer feedforward ReLU nets, we show that the global minimizer of a weakly-regularized cross-entropy loss is the max normalized margin solution among all neural nets, which generalizes well; ii) we develop a new technique for proving lower bounds for kernel methods, which relies on showing that the kernel cannot focus on informative features. Motivated by our generalization results, we study whether the regularized global optimum is attainable. We prove that for infinite-width two-layer nets, noisy gradient descent optimizes the regularized neural net loss to a global minimum in polynomial iterations.

stat.ML↗

arXiv2020

Med7: a transferable clinical natural language processing model for electronic health records

The field of clinical natural language processing has been advanced significantly since the introduction of deep learning models. The self-supervised representation learning and the transfer learning paradigm became the methods of choice in many natural language processing application, in particular in the settings with the dearth of high quality manually annotated data. Electronic health record systems are ubiquitous and the majority of patients' data are now being collected electronically and in particular in the form of free text. Identification of medical concepts and information extraction is a challenging task, yet important ingredient for parsing unstructured data into structured and tabulated format for downstream analytical tasks. In this work we introduced a named-entity recognition model for clinical natural language processing. The model is trained to recognise seven categories: drug names, route, frequency, dosage, strength, form, duration. The model was first self-supervisedly pre-trained by predicting the next word, using a collection of 2 million free-text patients' records from MIMIC-III corpora and then fine-tuned on the named-entity recognition task. The model achieved a lenient (strict) micro-averaged F1 score of 0.957 (0.893) across all seven categories. Additionally, we evaluated the transferability of the developed model using the data from the Intensive Care Unit in the US to secondary care mental health records (CRIS) in the UK. A direct application of the trained NER model to CRIS data resulted in reduced performance of F1=0.762, however after fine-tuning on a small sample from CRIS, the model achieved a reasonable performance of F1=0.944. This demonstrated that despite a close similarity between the data sets and the NER tasks, it is essential to fine-tune on the target domain data in order to achieve more accurate results.

cs.CL↗

arXiv2020

High-performance transmission of optical vortices through scattering media with opaque-lens field transformation theory

The applications of vortex beam in biomedical field are always hindered by light scattering of biological tissues. Recently wavefront shaping technology has shown the potential in focusing a structured light field through scattering media. Herein we have proposed a novel theory to describe the field transformation property of the opaque lens. This theory makes it possible to generate foci with specified amplitude and phase profiles by well-designing the incident field. We verify the validity of this theory by experimentally generating doughnut-shaped foci with various topological charges, and the foci match well with the theoretical predictions. Meanwhile, we have also analyzed the limitations of our approach, and it reveals that there is a trade-off between the thickness of the medium and the acceptance spatial bandwidth of the incident field. Our approach will inspire new research ideas for the deep-tissue applications of vortex beam in optics and dynamics, which is significant to numerous potential bio-photonics technologies.

physics.optics↗

arXiv2020

EdgeSlice: Slicing Wireless Edge Computing Network with Decentralized Deep Reinforcement Learning

5G and edge computing will serve various emerging use cases that have diverse requirements of multiple resources, e.g., radio, transportation, and computing. Network slicing is a promising technology for creating virtual networks that can be customized according to the requirements of different use cases. Provisioning network slices requires end-to-end resource orchestration which is challenging. In this paper, we design a decentralized resource orchestration system named EdgeSlice for dynamic end-to-end network slicing. EdgeSlice introduces a new decentralized deep reinforcement learning (D-DRL) method to efficiently orchestrate end-to-end resources. D-DRL is composed of a performance coordinator and multiple orchestration agents. The performance coordinator manages the resource orchestration policies in all the orchestration agents to ensure the service level agreement (SLA) of network slices. The orchestration agent learns the resource demands of network slices and orchestrates the resource allocation accordingly to optimize the performance of the slices under the constrained networking and computing resources. We design radio, transport and computing manager to enable dynamic configuration of end-to-end resources at runtime. We implement EdgeSlice on a prototype of the end-to-end wireless edge computing network with OpenAirInterface LTE network, OpenDayLight SDN switches, and CUDA GPU platform. The performance of EdgeSlice is evaluated through both experiments and trace-driven simulations. The evaluation results show that EdgeSlice achieves much improvement as compared to baseline in terms of performance, scalability, compatibility.

cs.NI↗