arXiv subjects

Yu Sun

Publications and source records attributed to Yu Sun.

At least 289 records · Page 16Linked to original sources

arXiv2020

Semi-Supervised Dialogue Policy Learning via Stochastic Reward Estimation

Dialogue policy optimization often obtains feedback until task completion in task-oriented dialogue systems. This is insufficient for training intermediate dialogue turns since supervision signals (or rewards) are only provided at the end of dialogues. To address this issue, reward learning has been introduced to learn from state-action pairs of an optimal policy to provide turn-by-turn rewards. This approach requires complete state-action annotations of human-to-human dialogues (i.e., expert demonstrations), which is labor intensive. To overcome this limitation, we propose a novel reward learning approach for semi-supervised policy learning. The proposed approach learns a dynamics model as the reward function which models dialogue progress (i.e., state-action sequences) based on expert demonstrations, either with or without annotations. The dynamics model computes rewards by predicting whether the dialogue progress is consistent with expert demonstrations. We further propose to learn action embeddings for a better generalization of the reward function. The proposed approach outperforms competitive policy learning baselines on MultiWOZ, a benchmark multi-domain dataset.

cs.CL↗

arXiv2020

Learning Individual Models for Imputation (Technical Report)

Missing numerical values are prevalent, e.g., owing to unreliable sensor reading, collection and transmission among heterogeneous sources. Unlike categorized data imputation over a limited domain, the numerical values suffer from two issues: (1) sparsity problem, the incomplete tuple may not have sufficient complete neighbors sharing the same/similar values for imputation, owing to the (almost) infinite domain; (2) heterogeneity problem, different tuples may not fit the same (regression) model. In this study, enlightened by the conditional dependencies that hold conditionally over certain tuples rather than the whole relation, we propose to learn a regression model individually for each complete tuple together with its neighbors. Our IIM, Imputation via Individual Models, thus no longer relies on sharing similar values among the k complete neighbors for imputation, but utilizes their regression results by the aforesaid learned individual (not necessary the same) models. Remarkably, we show that some existing methods are indeed special cases of our IIM, under the extreme settings of the number l of learning neighbors considered in individual learning. In this sense, a proper number l of neighbors is essential to learn the individual models (avoid over-fitting or under-fitting). We propose to adaptively learn individual models over various number l of neighbors for different complete tuples. By devising efficient incremental computation, the time complexity of learning a model reduces from linear to constant. Experiments on real data demonstrate that our IIM with adaptive learning achieves higher imputation accuracy than the existing approaches.

cs.DB↗

arXiv2020

First Investigation Into the Use of Deep Learning for Continuous Assessment of Neonatal Postoperative Pain

This paper presents the first investigation into the use of fully automated deep learning framework for assessing neonatal postoperative pain. It specifically investigates the use of Bilinear Convolutional Neural Network (B-CNN) to extract facial features during different levels of postoperative pain followed by modeling the temporal pattern using Recurrent Neural Network (RNN). Although acute and postoperative pain have some common characteristics (e.g., visual action units), postoperative pain has a different dynamic, and it evolves in a unique pattern over time. Our experimental results indicate a clear difference between the pattern of acute and postoperative pain. They also suggest the efficiency of using a combination of bilinear CNN with RNN model for the continuous assessment of postoperative pain intensity.

cs.CV↗

arXiv2020

mmWave Doubly-Massive-MIMO Communications Enhanced with an Intelligent Reflecting Surface

As a means to control wireless propagation environments, the use of emerging and novel intelligent reflecting surfaces (IRS) is envisioned to enhance and broaden many applications in future wireless networks. This paper is concerned with a point-to-point IRS-assisted millimeter-wave (mmWave) system in which the IRS consists of multiple subsurfaces, each having the same number of passive reflecting elements, whereas both the transmitter and receiver are equipped with massive antenna arrays. Under the scenario of having very large numbers of antennas at both transmit and receive ends, the achievable rate of the system is derived. Furthermore, with the objective of maximizing the achievable rate, the paper presents optimal solutions of power allocation, precoding/combining, and IRS's phase shifts. Then it is shown that when the number of reflecting elements at each subsurface is very large, the number of favorable and controllable propagation paths provided by the IRS is simply equal to the number of subsurfaces while the received signal-to-noise ratio corresponding to each of the favorable paths increases quadratically with the number of reflecting elements. In addition, the problem of minimizing the transmit power subject to the rate constraint is analyzed for the scenario without direct paths in the pure LOS propagation. Finally, numerical results are provided to corroborate the obtained analysis.

eess.SP↗

arXiv2020

Field-weighted Factorization Machines for Click-Through Rate Prediction in Display Advertising

Click-through rate (CTR) prediction is a critical task in online display advertising. The data involved in CTR prediction are typically multi-field categorical data, i.e., every feature is categorical and belongs to one and only one field. One of the interesting characteristics of such data is that features from one field often interact differently with features from different other fields. Recently, Field-aware Factorization Machines (FFMs) have been among the best performing models for CTR prediction by explicitly modeling such difference. However, the number of parameters in FFMs is in the order of feature number times field number, which is unacceptable in the real-world production systems. In this paper, we propose Field-weighted Factorization Machines (FwFMs) to model the different feature interactions between different fields in a much more memory-efficient way. Our experimental evaluations show that FwFMs can achieve competitive prediction performance with only as few as 4% parameters of FFMs. When using the same number of parameters, FwFMs can bring 0.92% and 0.47% AUC lift over FFMs on two real CTR prediction data sets.

cs.LG↗

arXiv2020

Predicting Different Types of Conversions with Multi-Task Learning in Online Advertising

Conversion prediction plays an important role in online advertising since Cost-Per-Action (CPA) has become one of the primary campaign performance objectives in the industry. Unlike click prediction, conversions have different types in nature, and each type may be associated with different decisive factors. In this paper, we formulate conversion prediction as a multi-task learning problem, so that the prediction models for different types of conversions can be learned together. These models share feature representations, but have their specific parameters, providing the benefit of information-sharing across all tasks. We then propose Multi-Task Field-weighted Factorization Machine (MT-FwFM) to solve these tasks jointly. Our experiment results show that, compared with two state-of-the-art models, MT-FwFM improve the AUC by 0.74% and 0.84% on two conversion types, and the weighted AUC across all conversion types is also improved by 0.50%.

cs.LG↗

arXiv2020

Boosting the Performance of Plug-and-Play Priors via Denoiser Scaling

Plug-and-play priors (PnP) is an image reconstruction framework that uses an image denoiser as an imaging prior. Unlike traditional regularized inversion, PnP does not require the prior to be expressible in the form of a regularization function. This flexibility enables PnP algorithms to exploit the most effective image denoisers, leading to their state-of-the-art performance in various imaging tasks. In this paper, we propose a new denoiser scaling technique to explicitly control the amount of PnP regularization. Traditionally, the performance of PnP algorithms is controlled via intrinsic parameters of the denoiser related to the noise variance. However, many powerful denoisers, such as the ones based on convolutional neural networks (CNNs), do not have tunable parameters that would allow controlling their influence within PnP. To address this issue, we introduce a scaling parameter that adjusts the magnitude of the denoiser input and output. We theoretical justify the denoiser scaling from the perspectives of proximal optimization, statistical estimation, and consensus equilibrium. Finally, we provide numerical experiments demonstrating the ability of denoiser scaling to systematically improve the performance of PnP for denoising CNN priors that do not have explicitly tunable parameters.

eess.IV↗

arXiv2020

Measurement of Liquid Flow Rate among the Annular Flow in Vertical Tee Junction

Since the liquid flow rate of the annular flow is closely related to the heat exchange efficiency, it has great significance to measure the liquid flow rate of the annular flow in vertical tee junction. In order to acquire the liquid flow rate of the annular flow in vertical tee junction, a measurement method has been designed, which implements the digital subtraction method to measure the thickness of the liquid film under the visible light and to apply the image feature matching algorithm to obtain the liquid velocity field. Moreover, the accuracy of the liquid film velocity field as well as the spatial and temporal stability of the mass flow rate is tested by proposed algorithms in this study. Experimental results show that the measurement error of our method is approximately 5% in the lower section of the main pipe and the branch pipe, and lower than 15% in the upper section of the main pipe. Therefore, this method has a high accuracy in comparison with other measurement approaches. Our method can be applied to measure and analyse the shape and property of the annular flow in the vertical tee junction.

physics.ins-det↗

arXiv2019

ERNIE 2.0: A Continual Pre-training Framework for Language Understanding

Recently, pre-trained models have achieved state-of-the-art results in various language understanding tasks, which indicates that pre-training on large-scale corpora may play a crucial role in natural language processing. Current pre-training procedures usually focus on training the model with several simple tasks to grasp the co-occurrence of words or sentences. However, besides co-occurring, there exists other valuable lexical, syntactic and semantic information in training corpora, such as named entity, semantic closeness and discourse relations. In order to extract to the fullest extent, the lexical, syntactic and semantic information from training corpora, we propose a continual pre-training framework named ERNIE 2.0 which builds and learns incrementally pre-training tasks through constant multi-task learning. Experimental results demonstrate that ERNIE 2.0 outperforms BERT and XLNet on 16 tasks including English tasks on GLUE benchmarks and several common tasks in Chinese. The source codes and pre-trained models have been released at https://github.com/PaddlePaddle/ERNIE.

cs.CL↗

arXiv2019

Indoor microbiome, environmental characteristics and asthma among junior high school students in Johor Bahru, Malaysia

Indoor microbial diversity and composition are suggested to affect the prevalence and severity of asthma. In this study, we collected floor dust and environmental characteristics from 21 classrooms, and health data related to asthma symptoms from 309 students, in junior high schools in Johor Bahru, Malaysia. Bacterial and fungal composition was characterized by sequencing 16s rRNA gene and internal transcribed spacer (ITS) region, and the absolute microbial concentration was quantified by qPCR. In total, 326 bacterial and 255 fungal genera were characterized. Five bacterial (Sphingobium, Rhodomicrobium, Shimwellia, Solirubrobacter, Pleurocapsa) and two fungal (Torulaspora and Leptosphaeriaceae) taxa were protective for asthma severity. Two bacterial taxa, Izhakiella and Robinsoniella, were positively associated with asthma severity. Several protective bacterial taxa including Rhodomicrobium, Shimwellia and Sphingobium has been reported as protective microbes in previous studies, whereas other taxa were first time reported. Environmental characteristics, such as age of building, size of textile curtain per room volume, occurrence of cockroaches, concentration of house dust mite allergens transferred from homes by the occupants, were involved in shaping the overall microbial community but not asthma-associated taxa; whereas visible dampness and mold, which did not change the overall microbial community for floor dust, decreased the concentration of protective bacteria Rhodomicrobium (\b{eta}=-2.86, p=0.021) of asthma, indicating complex interactions between microbes, environmental characteristics and asthma symptoms. Overall, this is the first indoor microbiome study to characterize the asthma-associated microbes and their environmental determinant in tropical area, promoting the understanding of microbial exposure and respiratory health in this region.

q-bio.GN↗

arXiv2019

Unsupervised Domain Adaptation through Self-Supervision

This paper addresses unsupervised domain adaptation, the setting where labeled training data is available on a source domain, but the goal is to have good performance on a target domain with only unlabeled data. Like much of previous work, we seek to align the learned representations of the source and target domains while preserving discriminability. The way we accomplish alignment is by learning to perform auxiliary self-supervised task(s) on both domains simultaneously. Each self-supervised task brings the two domains closer together along the direction relevant to that task. Training this jointly with the main task classifier on the source domain is shown to successfully generalize to the unlabeled target domain. The presented objective is straightforward to implement and easy to optimize. We achieve state-of-the-art results on four out of seven standard benchmarks, and competitive results on segmentation adaptation. We also demonstrate that our method composes well with another popular pixel-level adaptation method.

cs.LG↗

arXiv2019

Infusing Learned Priors into Model-Based Multispectral Imaging

We introduce a new algorithm for regularized reconstruction of multispectral (MS) images from noisy linear measurements. Unlike traditional approaches, the proposed algorithm regularizes the recovery problem by using a prior specified \emph{only} through a learned denoising function. More specifically, we propose a new accelerated gradient method (AGM) variant of regularization by denoising (RED) for model-based MS image reconstruction. The key ingredient of our approach is the three-dimensional (3D) deep neural net (DNN) denoiser that can fully leverage spationspectral correlations within MS images. Our results suggest the generalizability of our MS-RED algorithm, where a single trained DNN can be used to solve several different MS imaging problems.

eess.IV↗

arXiv2019

Block Coordinate Regularization by Denoising

We consider the problem of estimating a vector from its noisy measurements using a prior specified only through a denoising function. Recent work on plug-and-play priors (PnP) and regularization-by-denoising (RED) has shown the state-of-the-art performance of estimators under such priors in a range of imaging tasks. In this work, we develop a new block coordinate RED algorithm that decomposes a large-scale estimation problem into a sequence of updates over a small subset of the unknown variables. We theoretically analyze the convergence of the algorithm and discuss its relationship to the traditional proximal optimization. Our analysis complements and extends recent theoretical results for RED-based estimation methods. We numerically validate our method using several denoiser priors, including those based on convolutional neural network (CNN) denoisers.

cs.CV↗

arXiv2019

Harnessing the Power of Deep Learning Methods in Healthcare: Neonatal Pain Assessment from Crying Sound

Neonatal pain assessment in clinical environments is challenging as it is discontinuous and biased. Facial/body occlusion can occur in such settings due to clinical condition, developmental delays, prone position, or other external factors. In such cases, crying sound can be used to effectively assess neonatal pain. In this paper, we investigate the use of a novel CNN architecture (N-CNN) along with other CNN architectures (VGG16 and ResNet50) for assessing pain from crying sounds of neonates. The experimental results demonstrate that using our novel N-CNN for assessing pain from the sounds of neonates has a strong clinical potential and provides a viable alternative to the current assessment practice.

cs.CV↗

arXiv2019

Online Regularization by Denoising with Applications to Phase Retrieval

Regularization by denoising (RED) is a powerful framework for solving imaging inverse problems. Most RED algorithms are iterative batch procedures, which limits their applicability to very large datasets. In this paper, we address this limitation by introducing a novel online RED (On-RED) algorithm, which processes a small subset of the data at a time. We establish the theoretical convergence of On-RED in convex settings and empirically discuss its effectiveness in non-convex ones by illustrating its applicability to phase retrieval. Our results suggest that On-RED is an effective alternative to the traditional RED algorithms when dealing with large datasets.

eess.IV↗

arXiv2019

Multi-Channel Neural Network for Assessing Neonatal Pain from Videos

Neonates do not have the ability to either articulate pain or communicate it non-verbally by pointing. The current clinical standard for assessing neonatal pain is intermittent and highly subjective. This discontinuity and subjectivity can lead to inconsistent assessment, and therefore, inadequate treatment. In this paper, we propose a multi-channel deep learning framework for assessing neonatal pain from videos. The proposed framework integrates information from two pain indicators or channels, namely facial expression and body movement, using convolutional neural network (CNN). It also integrates temporal information using a recurrent neural network (LSTM). The experimental results prove the efficiency and superiority of the proposed temporal and multi-channel framework as compared to existing similar methods.

cs.CV↗

arXiv2019

RLTM: An Efficient Neural IR Framework for Long Documents

Deep neural networks have achieved significant improvements in information retrieval (IR). However, most existing models are computational costly and can not efficiently scale to long documents. This paper proposes a novel End-to-End neural ranking framework called Reinforced Long Text Matching (RLTM) which matches a query with long documents efficiently and effectively. The core idea behind the framework can be analogous to the human judgment process which firstly locates the relevance parts quickly from the whole document and then matches these parts with the query carefully to obtain the final label. Firstly, we select relevant sentences from the long documents by a coarse and efficient matching model. Secondly, we generate a relevance score by a more sophisticated matching model based on the sentence selected. The whole model is trained jointly with reinforcement learning in a pairwise manner by maximizing the expected score gaps between positive and negative examples. Experimental results demonstrate that RLTM has greatly improved the efficiency and effectiveness of the state-of-the-art models.

cs.CL↗

arXiv2019

CARL: Aggregated Search with Context-Aware Module Embedding Learning

Aggregated search aims to construct search result pages (SERPs) from blue-links and heterogeneous modules (such as news, images, and videos). Existing studies have largely ignored the correlations between blue-links and heterogeneous modules when selecting the heterogeneous modules to be presented. We observe that the top ranked blue-links, which we refer to as the \emph{context}, can provide important information about query intent and helps identify the relevant heterogeneous modules. For example, informative terms like "streamed" and "recorded" in the context imply that a video module may better satisfy the query. To model and utilize the context information for aggregated search, we propose a model with context attention and representation learning (CARL). Our model applies a recurrent neural network with an attention mechanism to encode the context, and incorporates the encoded context information into module embeddings. The context-aware module embeddings together with the ranking policy are jointly optimized under the Markov decision process (MDP) formulation. To achieve a more effective joint learning, we further propose an optimization function with self-supervision loss to provide auxiliary supervision signals. Experimental results based on two public datasets demonstrate the superiority of CARL over multiple baseline approaches, and confirm the effectiveness of the proposed optimization function in boosting the joint learning process.

cs.IR↗