arXiv subjects

Mu Li

Publications and source records attributed to Mu Li.

At least 109 records · Page 6Linked to original sources

arXiv2018

Achieving Human Parity on Automatic Chinese to English News Translation

Machine translation has made rapid advances in recent years. Millions of people are using it today in online translation systems and mobile applications in order to communicate across language barriers. The question naturally arises whether such systems can approach or achieve parity with human translations. In this paper, we first address the problem of how to define and accurately measure human parity in translation. We then describe Microsoft's machine translation system and measure the quality of its translations on the widely used WMT 2017 news translation task from Chinese to English. We find that our latest neural machine translation system has reached a new state-of-the-art, and that the translation quality is at human parity when compared to professional human translations. We also find that it significantly exceeds the quality of crowd-sourced non-professional translations.

cs.CL↗

arXiv2018

Shift-Net: Image Inpainting via Deep Feature Rearrangement

Deep convolutional networks (CNNs) have exhibited their potential in image inpainting for producing plausible results. However, in most existing methods, e.g., context encoder, the missing parts are predicted by propagating the surrounding convolutional features through a fully connected layer, which intends to produce semantically plausible but blurry result. In this paper, we introduce a special shift-connection layer to the U-Net architecture, namely Shift-Net, for filling in missing regions of any shape with sharp structures and fine-detailed textures. To this end, the encoder feature of the known region is shifted to serve as an estimation of the missing parts. A guidance loss is introduced on decoder feature to minimize the distance between the decoder feature after fully connected layer and the ground-truth encoder feature of the missing parts. With such constraint, the decoder feature in missing region can be used to guide the shift of encoder feature in known region. An end-to-end learning algorithm is further developed to train the Shift-Net. Experiments on the Paris StreetView and Places datasets demonstrate the efficiency and effectiveness of our Shift-Net in producing sharper, fine-detailed, and visually plausible results. The codes and pre-trained models are available at https://github.com/Zhaoyi-Yan/Shift-Net.

cs.CV↗

arXiv2018

Joint Training for Neural Machine Translation Models with Monolingual Data

Monolingual data have been demonstrated to be helpful in improving translation quality of both statistical machine translation (SMT) systems and neural machine translation (NMT) systems, especially in resource-poor or domain adaptation tasks where parallel data are not rich enough. In this paper, we propose a novel approach to better leveraging monolingual data for neural machine translation by jointly learning source-to-target and target-to-source NMT models for a language pair with a joint EM optimization method. The training process starts with two initial NMT models pre-trained on parallel data for each direction, and these two models are iteratively updated by incrementally decreasing translation losses on training data. In each iteration step, both NMT models are first used to translate monolingual data from one language to the other, forming pseudo-training data of the other NMT model. Then two new NMT models are learnt from parallel data together with the pseudo training data. Both NMT models are expected to be improved and better pseudo-training data can be generated in next step. Experiment results on Chinese-English and English-German translation tasks show that our approach can simultaneously improve translation quality of source-to-target and target-to-source models, significantly outperforming strong baseline systems which are enhanced with monolingual data for model training including back-translation.

cs.CL↗

arXiv2017

Evaluation of microstructure and mechanical property variations in AlxCoCrFeNi high entropy alloys produced by a high-throughput laser deposition method

Twenty-one distinct AlxCoCrFeNi alloys were rapidly prepared by laser alloying an equiatomic CoCrFeNi substrate with Al powder to create an alloy library ranging x=0.15-1.32. Variations in crystal structure, microstructure and mechanical properties were investigated using X-ray diffraction, scanning electron microscopy, scanning transmission electron microscopy and nanoindentation. With increasing Al content, the crystal structure transitioned from a disordered FCC to a mixture of disordered BCC and ordered B2 structures. While the onset of BCC/B2 formation was consistent with previously reported cast alloys, the FCC structure was observed at larger Al contents in the laser processed materials, resulting in a wider two phase regime. The FCC phase was primarily confined to the BCC/B2 grain boundaries at these high Al contents. The nanoindentation modulus and hardness of the FCC phase increased with Al content, while the properties of the BCC/B2 structure were insensitive to composition. The structure and mechanical properties of the laser-processed alloys were surprisingly consistent with reported results for cast alloys, demonstrating the feasibility of applying this high-throughput methodology to multicomponent alloy design.

cond-mat.mtrl-sci↗

arXiv2017

Learning Convolutional Networks for Content-weighted Image Compression

Lossy image compression is generally formulated as a joint rate-distortion optimization to learn encoder, quantizer, and decoder. However, the quantizer is non-differentiable, and discrete entropy estimation usually is required for rate control. These make it very challenging to develop a convolutional network (CNN)-based image compression system. In this paper, motivated by that the local information content is spatially variant in an image, we suggest that the bit rate of the different parts of the image should be adapted to local content. And the content aware bit rate is allocated under the guidance of a content-weighted importance map. Thus, the sum of the importance map can serve as a continuous alternative of discrete entropy estimation to control compression rate. And binarizer is adopted to quantize the output of encoder due to the binarization scheme is also directly defined by the importance map. Furthermore, a proxy function is introduced for binary operation in backward propagation to make it differentiable. Therefore, the encoder, decoder, binarizer and importance map can be jointly optimized in an end-to-end manner by using a subset of the ImageNet database. In low bit rate image compression, experiments show that our system significantly outperforms JPEG and JPEG 2000 by structural similarity (SSIM) index, and can produce the much better visual result with sharp edges, rich textures, and fewer artifacts.

cs.CV↗

arXiv2017

On the Powerball Method for Optimization

We propose a new method to accelerate the convergence of optimization algorithms. This method simply adds a power coefficient $γ\in[0,1)$ to the gradient during optimization. We call this the Powerball method and analyze the convergence rate for the Powerball method for strongly convex functions. While theoretically the Powerball method is guaranteed to have a linear convergence rate in the same order of the gradient method, we show that empirically it significantly outperforms the gradient descent and Newton's method, especially during the initial iterations. We demonstrate that the Powerball method provides a $10$-fold speedup of the convergence of both gradient descent and L-BFGS on multiple real datasets.

eess.SY↗

arXiv2016

Data Driven Resource Allocation for Distributed Learning

In distributed machine learning, data is dispatched to multiple machines for processing. Motivated by the fact that similar data points often belong to the same or similar classes, and more generally, classification rules of high accuracy tend to be "locally simple but globally complex" (Vapnik & Bottou 1993), we propose data dependent dispatching that takes advantage of such structure. We present an in-depth analysis of this model, providing new algorithms with provable worst-case guarantees, analysis proving existing scalable heuristics perform well in natural non worst-case conditions, and techniques for extending a dispatching rule from a small sample to the entire distribution. We overcome novel technical challenges to satisfy important conditions for accurate distributed learning, including fault tolerance and balancedness. We empirically compare our approach with baselines based on random partitioning, balanced partition trees, and locality sensitive hashing, showing that we achieve significantly higher accuracy on both synthetic and real world image and advertising datasets. We also demonstrate that our technique strongly scales with the available computing power.

cs.LG↗

arXiv2016

Convolutional Network for Attribute-driven and Identity-preserving Human Face Generation

This paper focuses on the problem of generating human face pictures from specific attributes. The existing CNN-based face generation models, however, either ignore the identity of the generated face or fail to preserve the identity of the reference face image. Here we address this problem from the view of optimization, and suggest an optimization model to generate human face with the given attributes while keeping the identity of the reference image. The attributes can be obtained from the attribute-guided image or by tuning the attribute features of the reference image. With the deep convolutional network "VGG-Face", the loss is defined on the convolutional feature maps. We then apply the gradient decent algorithm to solve this optimization problem. The results validate the effectiveness of our method for attribute driven and identity-preserving face generation.

cs.CV↗

arXiv2016

Revise Saturated Activation Functions

In this paper, we revise two commonly used saturated functions, the logistic sigmoid and the hyperbolic tangent (tanh). We point out that, besides the well-known non-zero centered property, slope of the activation function near the origin is another possible reason making training deep networks with the logistic function difficult to train. We demonstrate that, with proper rescaling, the logistic sigmoid achieves comparable results with tanh. Then following the same argument, we improve tahn by penalizing in the negative part. We show that "penalized tanh" is comparable and even outperforms the state-of-the-art non-saturated functions including ReLU and leaky ReLU on deep convolution neural networks. Our results contradict to the conclusion of previous works that the saturation property causes the slow convergence. It suggests further investigation is necessary to better understand activation functions in deep architectures.

cs.LG↗

arXiv2016

Length Scale of Correlated Dynamics in Ultra-thin Molecular Glasses

Physical vapor deposition (PVD) is widely used in manufacturing ultra-thin layers of amorphous organic solids. Here, we demonstrate that these films exhibit a sharp transition from glassy solid to liquid-like behavior with thickness below 30 nm. This liquid-like behavior persists even at temperatures well below the glass transition temperature, T$_{\mathrm{g}}$. The enhanced dynamics in these films can produce large scale morphological features during PVD and lead to a dewetting instability in films held at temperatures as low as T$_{\mathrm{g}}$-35 K. We measure the effective viscosity of organic glass films by monitoring the dewetting kinetics. These measurements combined with cooling rate-dependent T$_{\mathrm{g}}$ measurements show that the apparent activation barrier for rearrangement decreases sharply in films thinner than 30 nm. These observations suggest long-range facilitation of dynamics induced by the free surface, with dramatic effects on the properties of nano-scale amorphous materials.

cond-mat.soft↗

arXiv2016

Implicit Distortion and Fertility Models for Attention-based Encoder-Decoder NMT Model

Neural machine translation has shown very promising results lately. Most NMT models follow the encoder-decoder framework. To make encoder-decoder models more flexible, attention mechanism was introduced to machine translation and also other tasks like speech recognition and image captioning. We observe that the quality of translation by attention-based encoder-decoder can be significantly damaged when the alignment is incorrect. We attribute these problems to the lack of distortion and fertility models. Aiming to resolve these problems, we propose new variations of attention-based encoder-decoder and compare them with other models on machine translation. Our proposed method achieved an improvement of 2 BLEU points over the original attention-based encoder-decoder.

cs.CL↗

arXiv2015

MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems

MXNet is a multi-language machine learning (ML) library to ease the development of ML algorithms, especially for deep neural networks. Embedded in the host language, it blends declarative symbolic expression with imperative tensor computation. It offers auto differentiation to derive gradients. MXNet is computation and memory efficient and runs on various heterogeneous systems, ranging from mobile devices to distributed GPU clusters. This paper describes both the API design and the system implementation of MXNet, and explains how embedding of both symbolic expression and tensor operation is handled in a unified fashion. Our preliminary experiments reveal promising results on large scale deep neural network applications using multiple GPU machines.

cs.DC↗

arXiv2015

Empirical Evaluation of Rectified Activations in Convolutional Network

In this paper we investigate the performance of different types of rectified activation functions in convolutional neural network: standard rectified linear unit (ReLU), leaky rectified linear unit (Leaky ReLU), parametric rectified linear unit (PReLU) and a new randomized leaky rectified linear units (RReLU). We evaluate these activation function on standard image classification task. Our experiments suggest that incorporating a non-zero slope for negative part in rectified activation units could consistently improve the results. Thus our findings are negative on the common belief that sparsity is the key of good performance in ReLU. Moreover, on small scale dataset, using deterministic negative slope or learning it are both prone to overfitting. They are not as effective as using their randomized counterpart. By using RReLU, we achieved 75.68\% accuracy on CIFAR-100 test set without multiple test or ensemble.

cs.LG↗

arXiv2015

High Performance Latent Variable Models

Latent variable models have accumulated a considerable amount of interest from the industry and academia for their versatility in a wide range of applications. A large amount of effort has been made to develop systems that is able to extend the systems to a large scale, in the hope to make use of them on industry scale data. In this paper, we describe a system that operates at a scale orders of magnitude higher than previous works, and an order of magnitude faster than state-of-the-art system at the same scale, at the same time showing more robustness and more accurate results. Our system uses a number of advances in distributed inference: high performance in synchronization of sufficient statistics with relaxed consistency model; fast sampling, using the Metropolis-Hastings-Walker method to overcome dense generative models; statistical modeling, moving beyond Latent Dirichlet Allocation (LDA) to Pitman-Yor distributions (PDP) and Hierarchical Dirichlet Process (HDP) models; sophisticated parameter projection schemes, to resolve the conflicts within the constraint between parameters arising from the relaxed consistency model. This work significantly extends the domain of applicability of what is commonly known as the Parameter Server. We obtain results with up to hundreds billion oftokens, thousands of topics, and a vocabulary of a few million token-types, using up to 60,000 processor cores operating on a production cluster of a large Internet company. This demonstrates the feasibility to scale to problems orders of magnitude larger than any previously published work.

cs.LG↗

arXiv2015

AdaDelay: Delay Adaptive Distributed Stochastic Convex Optimization

We study distributed stochastic convex optimization under the delayed gradient model where the server nodes perform parameter updates, while the worker nodes compute stochastic gradients. We discuss, analyze, and experiment with a setup motivated by the behavior of real-world distributed computation networks, where the machines are differently slow at different time. Therefore, we allow the parameter updates to be sensitive to the actual delays experienced, rather than to worst-case bounds on the maximum delay. This sensitivity leads to larger stepsizes, that can help gain rapid initial convergence without having to wait too long for slower machines, while maintaining the same asymptotic complexity. We obtain encouraging improvements to overall convergence for distributed experiments on real datasets with up to billions of examples and features.

stat.ML↗

arXiv2015

Graph Partitioning via Parallel Submodular Approximation to Accelerate Distributed Machine Learning

Distributed computing excels at processing large scale data, but the communication cost for synchronizing the shared parameters may slow down the overall performance. Fortunately, the interactions between parameter and data in many problems are sparse, which admits efficient partition in order to reduce the communication overhead. In this paper, we formulate data placement as a graph partitioning problem. We propose a distributed partitioning algorithm. We give both theoretical guarantees and a highly efficient implementation. We also provide a highly efficient implementation of the algorithm and demonstrate its promising results on both text datasets and social networks. We show that the proposed algorithm leads to 1.6x speedup of a state-of-the-start distributed machine learning system by eliminating 90\% of the network communication.

cs.DC↗

arXiv2015

Beyond Word-based Language Model in Statistical Machine Translation

Language model is one of the most important modules in statistical machine translation and currently the word-based language model dominants this community. However, many translation models (e.g. phrase-based models) generate the target language sentences by rendering and compositing the phrases rather than the words. Thus, it is much more reasonable to model dependency between phrases, but few research work succeed in solving this problem. In this paper, we tackle this problem by designing a novel phrase-based language model which attempts to solve three key sub-problems: 1, how to define a phrase in language model; 2, how to determine the phrase boundary in the large-scale monolingual data in order to enlarge the training set; 3, how to alleviate the data sparsity problem due to the huge vocabulary size of phrases. By carefully handling these issues, the extensive experiments on Chinese-to-English translation show that our phrase-based language model can significantly improve the translation quality by up to +1.47 absolute BLEU score.

cs.CL↗

arXiv2014

Twofold Video Hashing with Automatic Synchronization

Video hashing finds a wide array of applications in content authentication, robust retrieval and anti-piracy search. While much of the existing research has focused on extracting robust and secure content descriptors, a significant open challenge still remains: Most existing video hashing methods are fallible to temporal desynchronization. That is, when the query video results by deleting or inserting some frames from the reference video, most existing methods assume the positions of the deleted (or inserted) frames are either perfectly known or reliably estimated. This assumption may be okay under typical transcoding and frame-rate changes but is highly inappropriate in adversarial scenarios such as anti-piracy video search. For example, an illegal uploader will try to bypass the 'piracy check' mechanism of YouTube/Dailymotion etc by performing a cleverly designed non-uniform resampling of the video. We present a new solution based on dynamic time warping (DTW), which can implement automatic synchronization and can be used together with existing video hashing methods. The second contribution of this paper is to propose a new robust feature extraction method called flow hashing (FH), based on frame averaging and optical flow descriptors. Finally, a fusion mechanism called distance boosting is proposed to combine the information extracted by DTW and FH. Experiments on real video collections show that such a hash extraction and comparison enables unprecedented robustness under both spatial and temporal attacks.

cs.MM↗