arXiv subjects

Lin Ma

Publications and source records attributed to Lin Ma.

At least 271 records · Page 15Linked to original sources

arXiv2018

Safe Element Screening for Submodular Function Minimization

Submodular functions are discrete analogs of convex functions, which have applications in various fields, including machine learning and computer vision. However, in large-scale applications, solving Submodular Function Minimization (SFM) problems remains challenging. In this paper, we make the first attempt to extend the emerging technique named screening in large-scale sparse learning to SFM for accelerating its optimization process. We first conduct a careful studying of the relationships between SFM and the corresponding convex proximal problems, as well as the accurate primal optimum estimation of the proximal problems. Relying on this study, we subsequently propose a novel safe screening method to quickly identify the elements guaranteed to be included (we refer to them as active) or excluded (inactive) in the final optimal solution of SFM during the optimization process. By removing the inactive elements and fixing the active ones, the problem size can be dramatically reduced, leading to great savings in the computational cost without sacrificing any accuracy. To the best of our knowledge, the proposed method is the first screening method in the fields of SFM and even combinatorial optimization, thus pointing out a new direction for accelerating SFM algorithms. Experiment results on both synthetic and real datasets demonstrate the significant speedups gained by our approach.

stat.ML↗

arXiv2018

Hierarchical Attention-Based Recurrent Highway Networks for Time Series Prediction

Time series prediction has been studied in a variety of domains. However, it is still challenging to predict future series given historical observations and past exogenous data. Existing methods either fail to consider the interactions among different components of exogenous variables which may affect the prediction accuracy, or cannot model the correlations between exogenous data and target data. Besides, the inherent temporal dynamics of exogenous data are also related to the target series prediction, and thus should be considered as well. To address these issues, we propose an end-to-end deep learning model, i.e., Hierarchical attention-based Recurrent Highway Network (HRHN), which incorporates spatio-temporal feature extraction of exogenous variables and temporal dynamics modeling of target variables into a single framework. Moreover, by introducing the hierarchical attention mechanism, HRHN can adaptively select the relevant exogenous features in different semantic levels. We carry out comprehensive empirical evaluations with various methods over several datasets, and show that HRHN outperforms the state of the arts in time series prediction, especially in capturing sudden changes and sudden oscillations of time series.

cs.LG↗

arXiv2018

Long-Term Human Motion Prediction by Modeling Motion Context and Enhancing Motion Dynamic

Human motion prediction aims at generating future frames of human motion based on an observed sequence of skeletons. Recent methods employ the latest hidden states of a recurrent neural network (RNN) to encode the historical skeletons, which can only address short-term prediction. In this work, we propose a motion context modeling by summarizing the historical human motion with respect to the current prediction. A modified highway unit (MHU) is proposed for efficiently eliminating motionless joints and estimating next pose given the motion context. Furthermore, we enhance the motion dynamic by minimizing the gram matrix loss for long-term motion prediction. Experimental results show that the proposed model can promisingly forecast the human future movements, which yields superior performances over related state-of-the-art approaches. Moreover, specifying the motion context with the activity labels enables our model to perform human motion transfer.

cs.CV↗

arXiv2018

Regularizing RNNs for Caption Generation by Reconstructing The Past with The Present

Recently, caption generation with an encoder-decoder framework has been extensively studied and applied in different domains, such as image captioning, code captioning, and so on. In this paper, we propose a novel architecture, namely Auto-Reconstructor Network (ARNet), which, coupling with the conventional encoder-decoder framework, works in an end-to-end fashion to generate captions. ARNet aims at reconstructing the previous hidden state with the present one, besides behaving as the input-dependent transition operator. Therefore, ARNet encourages the current hidden state to embed more information from the previous one, which can help regularize the transition dynamics of recurrent neural networks (RNNs). Extensive experimental results show that our proposed ARNet boosts the performance over the existing encoder-decoder models on both image captioning and source code captioning tasks. Additionally, ARNet remarkably reduces the discrepancy between training and inference processes for caption generation. Furthermore, the performance on permuted sequential MNIST demonstrates that ARNet can effectively regularize RNN, especially on modeling long-term dependencies. Our code is available at: https://github.com/chenxinpeng/ARNet

cs.CV↗

arXiv2018

Fine-grained Video Attractiveness Prediction Using Multimodal Deep Learning on a Large Real-world Dataset

Nowadays, billions of videos are online ready to be viewed and shared. Among an enormous volume of videos, some popular ones are widely viewed by online users while the majority attract little attention. Furthermore, within each video, different segments may attract significantly different numbers of views. This phenomenon leads to a challenging yet important problem, namely fine-grained video attractiveness prediction. However, one major obstacle for such a challenging problem is that no suitable benchmark dataset currently exists. To this end, we construct the first fine-grained video attractiveness dataset, which is collected from one of the most popular video websites in the world. In total, the constructed FVAD consists of 1,019 drama episodes with 780.6 hours covering different categories and a wide variety of video contents. Apart from the large amount of videos, hundreds of millions of user behaviors during watching videos are also included, such as "view counts", "fast-forward", "fast-rewind", and so on, where "view counts" reflects the video attractiveness while other engagements capture the interactions between the viewers and videos. First, we demonstrate that video attractiveness and different engagements present different relationships. Second, FVAD provides us an opportunity to study the fine-grained video attractiveness prediction problem. We design different sequential models to perform video attractiveness prediction by relying solely on video contents. The sequential models exploit the multimodal relationships between visual and audio components of the video contents at different levels. Experimental results demonstrate the effectiveness of our proposed sequential models with different visual and audio representations, the necessity of incorporating the two modalities, and the complementary behaviors of the sequential prediction models at different levels.

cs.CV↗

arXiv2018

Bidirectional Attentive Fusion with Context Gating for Dense Video Captioning

Dense video captioning is a newly emerging task that aims at both localizing and describing all events in a video. We identify and tackle two challenges on this task, namely, (1) how to utilize both past and future contexts for accurate event proposal predictions, and (2) how to construct informative input to the decoder for generating natural event descriptions. First, previous works predominantly generate temporal event proposals in the forward direction, which neglects future video context. We propose a bidirectional proposal method that effectively exploits both past and future contexts to make proposal predictions. Second, different events ending at (nearly) the same time are indistinguishable in the previous works, resulting in the same captions. We solve this problem by representing each event with an attentive fusion of hidden states from the proposal module and video contents (e.g., C3D features). We further propose a novel context gating mechanism to balance the contributions from the current event and its surrounding contexts dynamically. We empirically show that our attentively fused event representation is superior to the proposal hidden states or video contents alone. By coupling proposal and captioning modules into one unified framework, our model outperforms the state-of-the-arts on the ActivityNet Captions dataset with a relative gain of over 100% (Meteor score increases from 4.82 to 9.65).

cs.CV↗

arXiv2018

Learning to Guide Decoding for Image Captioning

Recently, much advance has been made in image captioning, and an encoder-decoder framework has achieved outstanding performance for this task. In this paper, we propose an extension of the encoder-decoder framework by adding a component called guiding network. The guiding network models the attribute properties of input images, and its output is leveraged to compose the input of the decoder at each time step. The guiding network can be plugged into the current encoder-decoder framework and trained in an end-to-end manner. Hence, the guiding vector can be adaptively learned according to the signal from the decoder, making itself to embed information from both image and language. Additionally, discriminative supervision can be employed to further improve the quality of guidance. The advantages of our proposed approach are verified by experiments carried out on the MS COCO dataset.

cs.CV↗

arXiv2018

Gated Fusion Network for Single Image Dehazing

In this paper, we propose an efficient algorithm to directly restore a clear image from a hazy input. The proposed algorithm hinges on an end-to-end trainable neural network that consists of an encoder and a decoder. The encoder is exploited to capture the context of the derived input images, while the decoder is employed to estimate the contribution of each input to the final dehazed result using the learned representations attributed to the encoder. The constructed network adopts a novel fusion-based strategy which derives three inputs from an original hazy image by applying White Balance (WB), Contrast Enhancing (CE), and Gamma Correction (GC). We compute pixel-wise confidence maps based on the appearance differences between these different inputs to blend the information of the derived inputs and preserve the regions with pleasant visibility. The final dehazed image is yielded by gating the important features of the derived inputs. To train the network, we introduce a multi-scale approach such that the halo artifacts can be avoided. Extensive experimental results on both synthetic and real-world images demonstrate that the proposed algorithm performs favorably against the state-of-the-art algorithms.

cs.CV↗

arXiv2018

Learning to Generate Time-Lapse Videos Using Multi-Stage Dynamic Generative Adversarial Networks

Taking a photo outside, can we predict the immediate future, e.g., how would the cloud move in the sky? We address this problem by presenting a generative adversarial network (GAN) based two-stage approach to generating realistic time-lapse videos of high resolution. Given the first frame, our model learns to generate long-term future frames. The first stage generates videos of realistic contents for each frame. The second stage refines the generated video from the first stage by enforcing it to be closer to real videos with regard to motion dynamics. To further encourage vivid motion in the final generated video, Gram matrix is employed to model the motion more precisely. We build a large scale time-lapse dataset, and test our approach on this new dataset. Using our model, we are able to generate realistic videos of up to $128\times 128$ resolution for 32 frames. Quantitative and qualitative experiment results have demonstrated the superiority of our model over the state-of-the-art models.

cs.CV↗

arXiv2018

Reconstruction Network for Video Captioning

In this paper, the problem of describing visual contents of a video sequence with natural language is addressed. Unlike previous video captioning work mainly exploiting the cues of video contents to make a language description, we propose a reconstruction network (RecNet) with a novel encoder-decoder-reconstructor architecture, which leverages both the forward (video to sentence) and backward (sentence to video) flows for video captioning. Specifically, the encoder-decoder makes use of the forward flow to produce the sentence description based on the encoded video semantic features. Two types of reconstructors are customized to employ the backward flow and reproduce the video features based on the hidden state sequence generated by the decoder. The generation loss yielded by the encoder-decoder and the reconstruction loss introduced by the reconstructor are jointly drawn into training the proposed RecNet in an end-to-end fashion. Experimental results on benchmark datasets demonstrate that the proposed reconstructor can boost the encoder-decoder models and leads to significant gains in video caption accuracy.

cs.CV↗

arXiv2018

Batch Auction Design For Cloud Container Services

Cloud containers represent a new, light-weight alternative to virtual machines in cloud computing. A user job may be described by a container graph that specifies the resource profile of each container and container dependence relations. This work is the first in the cloud computing literature that designs efficient market mechanisms for container based cloud jobs. Our design targets simultaneously incentive compatibility, computational efficiency, and economic efficiency. It further adapts the idea of batch online optimization into the paradigm of mechanism design, leveraging agile creation of cloud containers and exploiting delay tolerance of elastic cloud jobs. The new and classic techniques we employ include: (i) compact exponential optimization for expressing and handling non-traditional constraints that arise from container dependence and job deadlines; (ii) the primal-dual schema for designing efficient approximation algorithms for social welfare maximization; and (iii) posted price mechanisms for batch decision making and truthful payment design. Theoretical analysis and trace-driven empirical evaluation verify the efficacy of our container auction algorithms.

cs.GT↗

arXiv2017

Adaptive Neighboring Selection Algorithm Based on Curvature Prediction in Manifold Learning

Recently manifold learning algorithm for dimensionality reduction attracts more and more interests, and various linear and nonlinear, global and local algorithms are proposed. The key step of manifold learning algorithm is the neighboring region selection. However, so far for the references we know, few of which propose a generally accepted algorithm to well select the neighboring region. So in this paper, we propose an adaptive neighboring selection algorithm, which successfully applies the LLE and ISOMAP algorithms in the test. It is an algorithm that can find the optimal K nearest neighbors of the data points on the manifold. And the theoretical basis of the algorithm is the approximated curvature of the data point on the manifold. Based on Riemann Geometry, Jacob matrix is a proper mathematical concept to predict the approximated curvature. By verifying the proposed algorithm on embedding Swiss roll from R3 to R2 based on LLE and ISOMAP algorithm, the simulation results show that the proposed adaptive neighboring selection algorithm is feasible and able to find the optimal value of K, making the residual variance relatively small and better visualization of the results. By quantitative analysis, the embedding quality measured by residual variance is increased 45.45% after using the proposed algorithm in LLE.

stat.ME↗

arXiv2017

Joint Semi-supervised RSS Dimensionality Reduction and Fingerprint Based Algorithm for Indoor Localization

With the recent development in mobile computing devices and as the ubiquitous deployment of access points(APs) of Wireless Local Area Networks(WLANs), WLAN based indoor localization systems(WILSs) are of mounting concentration and are becoming more and more prevalent for they do not require additional infrastructure. As to the localization methods in WILSs, for the approaches used to localization in satellite based global position systems are difficult to achieve in indoor environments, fingerprint based localization algorithms(FLAs) are predominant in the RSS based schemes. However, the performance of FLAs has close relationship with the number of APs and the number of reference points(RPs) in WILSs, especially as the redundant deployment of APs and RPs in the system. There are two fatal problems, curse of dimensionality (CoD) and asymmetric matching(AM), caused by increasing number of APs and breaking down APs during online stage. In this paper, a semi-supervised RSS dimensionality reduction algorithm is proposed to solve these two dilemmas at the same time and there are numerous analyses about the theoretical realization of the proposed method. Another significant innovation of this paper is jointing the fingerprint based algorithm with CM-SDE algorithm to improve the localization accuracy of indoor localization.

stat.ML↗

arXiv2016

Stable artificial solid electrolyte interfaces for lithium batteries

A rechargeable lithium metal battery (LMB), which uses metallic lithium at the anode, is among the most promising technologies for next generation electrochemical energy storage devices due to its high energy density, particularly when Li is paired with energetic conversion cathodes such as sulfur, oxygen/air, and carbon dioxide. Practical LMBs in any of these designs remain elusive due to multiple stubborn problems, including parasitic reactions of Li metal with liquid electrolytes, unstable/dendritic electrodeposition at the anode during cell recharge, and chemical reaction of dissolved cathode conversion products with the Li anode. The solid electrolyte interface (SEI) formed between lithium metal and liquid electrolytes plays a critical role in all of these processes. We report on the chemistry and interfacial properties of artificial SEI films created by in-situ reaction of a strong Lewis Acid AlI3 additive, Li metal, and aprotic liquid electrolytes. We find that these SEI films impart exceptional interfacial stability to a Li metal anode. We further show that the improvements come from at least three processes: (i) in-situ formation of Li-Al alloy, (ii) formation of a LiI salt layer on Li, and (iii) creation of a stable polymer thin film on the lithium metal anode.

cond-mat.mtrl-sci↗

arXiv2016

Fabricating Multifunctional Nanoparticle Membranes by a Fast Layer-by-Layer Langmuir-Blodgett Process: Application in Lithium-sulfur Batteries

The Langmuir-Blodgett (LB) technique is a powerful, widely used method for preparing coatings of amphiphilic molecules at air/water interfaces with thickness control down to a single molecule. Here we report two new LB techniques designed to create ordered, multifunctional nanoparticle films on any non-reactive support. The methods utilize Marangoni stresses produced by surfactants at a fluid/solid/gas interface and self-assembly of nanoparticles to facilitate rapid creation of dense monolayers of multi-wall carbon nanotubes (MWCNT), metal-oxide nanoparticles, polymers, and combinations of these materials in a layer-by-layer configuration. Using the polyolefin separator in a lithium sulfur (Li-S) electrochemical cell as an example, we illustrate how the method can be used to create structured membranes for regulating mass and charge transport. We show that a layered MWCNT/SiO2/MWCNT nanomaterial created in a clip-like configuration, with gravimetric areal coverage of ~130 mg cm-2 and a thickness of ~3 micron, efficiently adsorbs dissolved lithium polysulfide (LiPS) species and efficiently reutilize them for improving Li-S battery performance.

physics.chem-ph↗

arXiv2015

Learning to Answer Questions From Image Using Convolutional Neural Network

In this paper, we propose to employ the convolutional neural network (CNN) for the image question answering (QA). Our proposed CNN provides an end-to-end framework with convolutional architectures for learning not only the image and question representations, but also their inter-modal interactions to produce the answer. More specifically, our model consists of three CNNs: one image CNN to encode the image content, one sentence CNN to compose the words of the question, and one multimodal convolution layer to learn their joint representation for the classification in the space of candidate answer words. We demonstrate the efficacy of our proposed model on the DAQUAR and COCO-QA datasets, which are two benchmark datasets for the image QA, with the performances significantly outperforming the state-of-the-art.

cs.CL↗

arXiv2015

Multimodal Convolutional Neural Networks for Matching Image and Sentence

In this paper, we propose multimodal convolutional neural networks (m-CNNs) for matching image and sentence. Our m-CNN provides an end-to-end framework with convolutional architectures to exploit image representation, word composition, and the matching relations between the two modalities. More specifically, it consists of one image CNN encoding the image content, and one matching CNN learning the joint representation of image and sentence. The matching CNN composes words to different semantic fragments and learns the inter-modal relations between image and the composed fragments at different levels, thus fully exploit the matching relations between image and sentence. Experimental results on benchmark databases of bidirectional image and sentence retrieval demonstrate that the proposed m-CNNs can effectively capture the information necessary for image and sentence matching. Specifically, our proposed m-CNNs for bidirectional image and sentence retrieval on Flickr30K and Microsoft COCO databases achieve the state-of-the-art performances.

cs.CV↗

arXiv2015

Estimating the Algorithmic Complexity of Stock Markets

Randomness and regularities in Finance are usually treated in probabilistic terms. In this paper, we develop a completely different approach in using a non-probabilistic framework based on the algorithmic information theory initially developed by Kolmogorov (1965). We present some elements of this theory and show why it is particularly relevant to Finance, and potentially to other sub-fields of Economics as well. We develop a generic method to estimate the Kolmogorov complexity of numeric series. This approach is based on an iterative "regularity erasing procedure" implemented to use lossless compression algorithms on financial data. Examples are provided with both simulated and real-world financial time series. The contributions of this article are twofold. The first one is methodological : we show that some structural regularities, invisible with classical statistical tests, can be detected by this algorithmic method. The second one consists in illustrations on the daily Dow-Jones Index suggesting that beyond several well-known regularities, hidden structure may in this index remain to be identified.

q-fin.CP↗