arXiv subjects

Ming Zhang

Publications and source records attributed to Ming Zhang.

At least 397 records · Page 22Linked to original sources

arXiv2018

Theory of Generative Deep Learning : Probe Landscape of Empirical Error via Norm Based Capacity Control

Despite its remarkable empirical success as a highly competitive branch of artificial intelligence, deep learning is often blamed for its widely known low interpretation and lack of firm and rigorous mathematical foundation. However, most theoretical endeavor is devoted in discriminative deep learning case, whose complementary part is generative deep learning. To the best of our knowledge, we firstly highlight landscape of empirical error in generative case to complete the full picture through exquisite design of image super resolution under norm based capacity control. Our theoretical advance in interpretation of the training dynamic is achieved from both mathematical and biological sides.

cs.LG↗

arXiv2018

An Online Algorithm for Power-proportional Data Centers with Switching Cost

Recent studies have shown that power-proportional data centers can save energy cost by dynamically "right-sizing" the data centers based on real-time workload. More servers are activated when the workload increases while some servers can be put into the sleep mode during periods of low load. In this paper, we revisit the dynamic right-sizing problem for heterogeneous data centers with various operational cost and switching cost. We propose a new online algorithm based on a regularization technique, which achieves a better competitive ratio compared to the state-of-the-art greedy algorithm. We further introduce a switching cost offset into the model and extend our algorithm to this new setting. Simulations based on real workload and renewable energy traces show that our algorithms outperform the greedy algorithm in both settings.

cs.NI↗

arXiv2018

Generation of multiphoton entangled quantum states with a single silicon nanowire

Multiphoton entanglement plays a critical role in quantum information processing, and greatly improves our fundamental understanding of the quantum world. Despite tremendous efforts in either bulk media or fiber-based devices, nonlinear interactions in integrated circuits show great promise as an excellent platform for photon pair generation with its high brightness, stability and scalability \cite{Caspani2017}. Here, we demonstrate the generation of bi- and multiphoton polarization entangled qubits in a single silicon nanowire waveguide, and these qubits directly compatible with the dense wavelength division multiplexing in telecommunication system. Multiphoton interference and quantum state tomography were used to characterize the quality of the entangled states. Four-photon entanglement states among two frequency channels were ascertained with a fidelity of $0.78\pm0.02$. Our work realizes the integrated multiphoton source in a relatively simple pattern and paves a way for the revolution of multiphoton quantum science.

quant-ph↗

arXiv2017

Towards Automated ICD Coding Using Deep Learning

International Classification of Diseases(ICD) is an authoritative health care classification system of different diseases and conditions for clinical and management purposes. Considering the complicated and dedicated process to assign correct codes to each patient admission based on overall diagnosis, we propose a hierarchical deep learning model with attention mechanism which can automatically assign ICD diagnostic codes given written diagnosis. We utilize character-aware neural language models to generate hidden representations of written diagnosis descriptions and ICD codes, and design an attention mechanism to address the mismatch between the numbers of descriptions and corresponding codes. Our experimental results show the strong potential of automated ICD coding from diagnosis descriptions. Our best model achieves 0.53 and 0.90 of F1 score and area under curve of receiver operating characteristic respectively. The result outperforms those achieved using character-unaware encoding method or without attention mechanism. It indicates that our proposed deep learning model can code automatically in a reasonable way and provide a framework for computer-auxiliary ICD coding.

cs.CL↗

arXiv2017

Coexistence curve and molecule number density of AdS topological charged black hole in massive gravity

The coexistence curve and molecule number density of a 4-dimensional AdS topological charged black hole in massive gravity is investigated. We find that the analytic expression of the coexistence curve in the reduced parameter space is dependent on theory parameters. This is very different from the previous results obtained in other modified gravity such as $f(R)$ gravity and Gauss-Bonnet gravity. Besides, we derive the explicit expression of the physical quantity which describes the difference of the number densities of AdS topological charged black hole molecules between the small and large black hole. It is observed that the difference of the molecule number densities is also dependent on theory parameters. Both the expressions of the coexistence curve and the difference of the molecule number densities can be reduced into a form which is similar to a RN-AdS black hole if the mass of graviton $m$ is zero. Moreover, we find the shifted temperature under massive gravity. This can highlight the important role played by the mass of graviton and other parameters in the phase transitions of AdS black holes in massive gravity.

hep-th↗

arXiv2017

Reentrant phase transitions and triple points of topological AdS black holes in Born-Infeld-massive gravity

Motivated by recent developments of black hole thermodynamics in de Rham, Gabadadze and Tolley(dRGT) massive gravity, we study the critical behaviors of four-dimensional topological Anti-de Sitter(AdS) black holes in the presence of Born-Infeld nonlinear electrodynamics by treating the cosmological constant as pressure and the corresponding conjugate quantity is interpreted as thermodynamic volume. It shows that besides the Van der Waals-like SBH/LBH phase transitions appears, the so-called reentrant phase transitions (RPTs) are also observed when the coupling coefficients $c_i m^2$ of massive potential and Born-Infeld parameter $b$ satisfy some certain conditions.

hep-th↗

arXiv2017

An Attention-based Collaboration Framework for Multi-View Network Representation Learning

Learning distributed node representations in networks has been attracting increasing attention recently due to its effectiveness in a variety of applications. Existing approaches usually study networks with a single type of proximity between nodes, which defines a single view of a network. However, in reality there usually exists multiple types of proximities between nodes, yielding networks with multiple views. This paper studies learning node representations for networks with multiple views, which aims to infer robust node representations across different views. We propose a multi-view representation learning approach, which promotes the collaboration of different views and lets them vote for the robust representations. During the voting process, an attention mechanism is introduced, which enables each node to focus on the most informative views. Experimental results on real-world networks show that the proposed approach outperforms existing state-of-the-art approaches for network representation learning with a single view and other competitive approaches with multiple views.

cs.SI↗

arXiv2017

IOTune: A G-states Driver for Elastic Performance of Block Storage

Imagining a disk which provides baseline performance at a relatively low price during low-load periods, but when workloads demand more resources, the disk performance is automatically promoted in situ and in real time. In a hardware era, this is hardly achievable. However, this imagined disk is becoming reality due to the technical advances of software-defined storage, which enable volume performance to be adjusted on the fly. We propose IOTune, a resource management middleware which employs software-defined storage primitives to implement G-states of virtual block devices. G-states enable virtual block devices to serve at multiple performance gears, getting rid of conflicts between immutable resource reservation and dynamic resource demands, and always achieving resource right-provisioning for workloads. Accompanying G-states, we also propose a new block storage pricing policy for cloud providers. Our case study for applying G-states to cloud block storage verifies the effectiveness of the IOTune framework. Trace-replay based evaluations demonstrate that storage volumes with G-states adapt to workload fluctuations. For tenants, G-states enable volumes to provide much better QoS with a same cost of ownership, comparing with static IOPS provisioning and the I/O credit mechanism. G-states also reduce I/O tail latencies by one to two orders of magnitude. From the standpoint of cloud providers, G-states promote storage utilization, creating values and benefiting competitiveness. G-states supported by IOTune provide a new paradigm for storage resource management and pricing in multi-tenant clouds.

cs.OS↗

arXiv2017

Reentrant phase transitions of higher-dimensional AdS black holes in dRGT massive gravity

We study the $P-V$ criticality and phase transition in the extended phase space of anti-de Sitter (AdS) black holes in higher-dimensional de Rham, Gabadadze and Tolley (dRGT) massive gravity, treating the cosmological constant as pressure and the corresponding conjugate quantity is interpreted as thermodynamic volume. Besides the usual small/large black hole phase transitions, the interesting thermodynamic phenomena of reentrant phase transitions (RPTs) are observed for black holes in all $d\geq6$-dimensional spacetime when the coupling coefficients $c_i m^2$ of massive potential satisfy some certain conditions.

gr-qc↗

arXiv2017

Syntax Aware LSTM Model for Chinese Semantic Role Labeling

As for semantic role labeling (SRL) task, when it comes to utilizing parsing information, both traditional methods and recent recurrent neural network (RNN) based methods use the feature engineering way. In this paper, we propose Syntax Aware Long Short Time Memory(SA-LSTM). The structure of SA-LSTM modifies according to dependency parsing information in order to model parsing information directly in an architecture engineering way instead of feature engineering way. We experimentally demonstrate that SA-LSTM gains more improvement from the model architecture. Furthermore, SA-LSTM outperforms the state-of-the-art on CPB 1.0 significantly according to Student t-test ($p<0.05$).

cs.CL↗

arXiv2017

Two-Bit Networks for Deep Learning on Resource-Constrained Embedded Devices

With the rapid proliferation of Internet of Things and intelligent edge devices, there is an increasing need for implementing machine learning algorithms, including deep learning, on resource-constrained mobile embedded devices with limited memory and computation power. Typical large Convolutional Neural Networks (CNNs) need large amounts of memory and computational power, and cannot be deployed on embedded devices efficiently. We present Two-Bit Networks (TBNs) for model compression of CNNs with edge weights constrained to (-2, -1, 1, 2), which can be encoded with two bits. Our approach can reduce the memory usage and improve computational efficiency significantly while achieving good performance in terms of classification accuracy, thus representing a reasonable tradeoff between model size and performance.

cs.LG↗

arXiv2016

Less is More: Learning Prominent and Diverse Topics for Data Summarization

Statistical topic models efficiently facilitate the exploration of large-scale data sets. Many models have been developed and broadly used to summarize the semantic structure in news, science, social media, and digital humanities. However, a common and practical objective in data exploration tasks is not to enumerate all existing topics, but to quickly extract representative ones that broadly cover the content of the corpus, i.e., a few topics that serve as a good summary of the data. Most existing topic models fit exactly the same number of topics as a user specifies, which have imposed an unnecessary burden to the users who have limited prior knowledge. We instead propose new models that are able to learn fewer but more representative topics for the purpose of data summarization. We propose a reinforced random walk that allows prominent topics to absorb tokens from similar and smaller topics, thus enhances the diversity among the top topics extracted. With this reinforced random walk as a general process embedded in classical topic models, we obtain \textit{diverse topic models} that are able to extract the most prominent and diverse topics from data. The inference procedures of these diverse topic models remain as simple and efficient as the classical models. Experimental results demonstrate that the diverse topic models not only discover topics that better summarize the data, but also require minimal prior knowledge of the users.

cs.LG↗

arXiv2016

Context-aware Natural Language Generation with Recurrent Neural Networks

This paper studied generating natural languages at particular contexts or situations. We proposed two novel approaches which encode the contexts into a continuous semantic representation and then decode the semantic representation into text sequences with recurrent neural networks. During decoding, the context information are attended through a gating mechanism, addressing the problem of long-range dependency caused by lengthy sequences. We evaluate the effectiveness of the proposed approaches on user review data, in which rich contexts are available and two informative contexts, sentiments and products, are selected for evaluation. Experiments show that the fake reviews generated by our approaches are very natural. Results of fake review detection with human judges show that more than 50\% of the fake reviews are misclassified as the real reviews, and more than 90\% are misclassified by existing state-of-the-art fake review detection algorithm.

cs.CL↗

arXiv2016

Improving Color Constancy by Discounting the Variation of Camera Spectral Sensitivity

It is an ill-posed problem to recover the true scene colors from a color biased image by discounting the effects of scene illuminant and camera spectral sensitivity (CSS) at the same time. Most color constancy (CC) models have been designed to first estimate the illuminant color, which is then removed from the color biased image to obtain an image taken under white light, without the explicit consideration of CSS effect on CC. This paper first studies the CSS effect on illuminant estimation arising in the inter-dataset-based CC (inter-CC), i.e., training a CC model on one dataset and then testing on another dataset captured by a distinct CSS. We show the clear degradation of existing CC models for inter-CC application. Then a simple way is proposed to overcome such degradation by first learning quickly a transform matrix between the two distinct CSSs (CSS-1 and CSS-2). The learned matrix is then used to convert the data (including the illuminant ground truth and the color biased images) rendered under CSS-1 into CSS-2, and then train and apply the CC model on the color biased images under CSS-2, without the need of burdensome acquiring of training set under CSS-2. Extensive experiments on synthetic and real images show that our method can clearly improve the inter-CC performance for traditional CC algorithms. We suggest that by taking the CSS effect into account, it is more likely to obtain the truly color constant images invariant to the changes of both illuminant and camera sensors.

cs.CV↗

arXiv2016

Two are Better than One: An Ensemble of Retrieval- and Generation-Based Dialog Systems

Open-domain human-computer conversation has attracted much attention in the field of NLP. Contrary to rule- or template-based domain-specific dialog systems, open-domain conversation usually requires data-driven approaches, which can be roughly divided into two categories: retrieval-based and generation-based systems. Retrieval systems search a user-issued utterance (called a query) in a large database, and return a reply that best matches the query. Generative approaches, typically based on recurrent neural networks (RNNs), can synthesize new replies, but they suffer from the problem of generating short, meaningless utterances. In this paper, we propose a novel ensemble of retrieval-based and generation-based dialog systems in the open domain. In our approach, the retrieved candidate, in addition to the original query, is fed to an RNN-based reply generator, so that the neural model is aware of more information. The generated reply is then fed back as a new candidate for post-reranking. Experimental results show that such ensemble outperforms each single part of it by a large margin.

cs.CL↗

arXiv2016

Dialogue Session Segmentation by Embedding-Enhanced TextTiling

In human-computer conversation systems, the context of a user-issued utterance is particularly important because it provides useful background information of the conversation. However, it is unwise to track all previous utterances in the current session as not all of them are equally important. In this paper, we address the problem of session segmentation. We propose an embedding-enhanced TextTiling approach, inspired by the observation that conversation utterances are highly noisy, and that word embeddings provide a robust way of capturing semantics. Experimental results show that our approach achieves better performance than the TextTiling, MMD approaches.

cs.CL↗

arXiv2016

Coexistent physics of massive black holes in the phase transitions

The coexistent physics of de Rham-Gabada-dze-Tolley (dRGT) massive black holes and holographic massive black holes is investigated in the extended phase space where the cosmological constant is viewed as pressure. Van der Waals like phase transitions are found for both of them. Coexistent curves of reduced pressure and reduced temperature are found to be different from that of RN-AdS black holes. Coexistent curves of reduced Gibbs free energy and reduced pressure show that Gibbs free energy in the canonical ensemble decreases monotonically with the increasing pressure. The concept number density is introduced to study the coexistent physics. It is uncovered that with the increasing pressure, the number densities of small black holes (SBHs) and large black holes (LBHs) change monotonically in the contrary directions till finally reaching the same value at the critical points of the phase transitions. In other words, with the increasing pressure the number density differences between SBHs and LBHs decrease monotonically before disappearance at the critical points. Further more, HEPM thermodynamic scalar curvature differences between SBHs and LBHs are found to decrease monotonically to zero when approaching to the critical points, which is similar as a RN-AdS black hole. We propose that both the number density difference and the HEPM scalar curvature difference can be order parameters describing the SBH/LBH phase transition and judging the upcoming of critical point where a second-order phase transition takes place. These results provide us with new recognition of the massive gravity. The thermodynamics in the extended phase space of AdS black holes is enriched.

gr-qc↗

arXiv2016

World Knowledge as Indirect Supervision for Document Clustering

One of the key obstacles in making learning protocols realistic in applications is the need to supervise them, a costly process that often requires hiring domain experts. We consider the framework to use the world knowledge as indirect supervision. World knowledge is general-purpose knowledge, which is not designed for any specific domain. Then the key challenges are how to adapt the world knowledge to domains and how to represent it for learning. In this paper, we provide an example of using world knowledge for domain dependent document clustering. We provide three ways to specify the world knowledge to domains by resolving the ambiguity of the entities and their types, and represent the data with world knowledge as a heterogeneous information network. Then we propose a clustering algorithm that can cluster multiple types and incorporate the sub-type information as constraints. In the experiments, we use two existing knowledge bases as our sources of world knowledge. One is Freebase, which is collaboratively collected knowledge about entities and their organizations. The other is YAGO2, a knowledge base automatically extracted from Wikipedia and maps knowledge to the linguistic knowledge base, WordNet. Experimental results on two text benchmark datasets (20newsgroups and RCV1) show that incorporating world knowledge as indirect supervision can significantly outperform the state-of-the-art clustering algorithms as well as clustering algorithms enhanced with world knowledge features.

cs.LG↗