arXiv subjects

Weiwei Cai

Publications and source records attributed to Weiwei Cai.

29 records · Page 2Linked to original sources

arXiv2020

Real Differences between OT and CRDT in Correctness and Complexity for Consistency Maintenance in Co-Editors

OT (Operational Transformation) was invented for supporting real-time co-editors in the late 1980s and has evolved to become core techniques widely used in today's working co-editors and adopted in industrial products. CRDT (Commutative Replicated Data Type) for co-editors was first proposed around 2006, under the name of WOOT (WithOut Operational Transformation). Follow-up CRDT variations are commonly labeled as "post-OT" techniques capable of making concurrent operations natively commutative in co-editors. On top of that, CRDT solutions have made broad claims of superiority over OT solutions, and often portrayed OT as an incorrect and inefficient technique. Over one decade later, however, CRDT is rarely found in working co-editors; OT remains the choice for building the vast majority of today's co-editors. Contradictions between the reality and CRDT's purported advantages have been the source of much confusion and debate among co-editing researcher sand developers. To seek truth from facts, we set out to conduct a comprehensive and critical review on representative OT and CRDT solutions and co-editors based on them. From this work, we have made important discoveries about OT and CRDT, and revealed facts and evidences that refute CRDT claims over OT on all accounts. These discoveries help explain the underlying reasons for the choice between OT and CRDT in the real world. We report these results in a series of three articles. In the second article of this series, we reveal the differences between OT and CRDT in their basic approaches to realizing the same general transformation and how such differences had resulted in different challenges and consequential correctness and complexity issues. Moreover, we reveal hidden complexity and algorithmic flaws with representative CRDT solutions, and discuss common myths and facts related to correctness and complexity of OT and CRDT.

cs.DC↗

arXiv2020

GraftNet: An Engineering Implementation of CNN for Fine-grained Multi-label Task

Multi-label networks with branches are proved to perform well in both accuracy and speed, but lacks flexibility in providing dynamic extension onto new labels due to the low efficiency of re-work on annotating and training. For multi-label classification task, to cover new labels we need to annotate not only newly collected images, but also the previous whole dataset to check presence of these new labels. Also training on whole re-annotated dataset costs much time. In order to recognize new labels more effectively and accurately, we propose GraftNet, which is a customizable tree-like network with its trunk pretrained with a dynamic graph for generic feature extraction, and branches separately trained on sub-datasets with single label to improve accuracy. GraftNet could reduce cost, increase flexibility, and incrementally handle new labels. Experimental results show that it has good performance on our human attributes recognition task, which is fine-grained multi-label classification.

cs.CV↗

arXiv2020

PiiGAN: Generative Adversarial Networks for Pluralistic Image Inpainting

The latest methods based on deep learning have achieved amazing results regarding the complex work of inpainting large missing areas in an image. But this type of method generally attempts to generate one single "optimal" result, ignoring many other plausible results. Considering the uncertainty of the inpainting task, one sole result can hardly be regarded as a desired regeneration of the missing area. In view of this weakness, which is related to the design of the previous algorithms, we propose a novel deep generative model equipped with a brand new style extractor which can extract the style feature (latent vector) from the ground truth. Once obtained, the extracted style feature and the ground truth are both input into the generator. We also craft a consistency loss that guides the generated image to approximate the ground truth. After iterations, our generator is able to learn the mapping of styles corresponding to multiple sets of vectors. The proposed model can generate a large number of results consistent with the context semantics of the image. Moreover, we evaluated the effectiveness of our model on three datasets, i.e., CelebA, PlantVillage, and MauFlex. Compared to state-of-the-art inpainting methods, this model is able to offer desirable inpainting results with both better quality and higher diversity. The code and model will be made available on https://github.com/vivitsai/PiiGAN.

cs.CV↗

arXiv2018

Real Differences between OT and CRDT for Co-Editors

OT (Operational Transformation) was invented for supporting real-time co-editors in the late 1980s and has evolved to become a core technique used in today's working co-editors and adopted in major industrial products. CRDT (Commutative Replicated Data Type) for co-editors was first proposed around 2006, under the name of WOOT (WithOut Operational Transformation). Follow-up CRDT variations are commonly labeled as "post-OT" techniques capable of making concurrent operations natively commutative, and have made broad claims of superiority over OT solutions, in terms of correctness, time and space complexity, simplicity, etc. Over one decade later, however, CRDT solutions are rarely found in working co-editors, while OT solutions remain the choice for building the vast majority of co-editors. Contradictions between this reality and CRDT's purported advantages have been the source of much debate and confusion in co-editing research and developer communities. What is CRDT really to co-editing? What are the real differences between OT and CRDT for co-editors? What are the key factors that may have affected the adoption of and choice between OT and CRDT for co-editors in the real world? In this paper, we report our discoveries, in relation to these questions and beyond, from a comprehensive review and comparison study on representative OT and CRDT solutions and working co-editors based on them. Moreover, this work reveals facts and presents evidences that refute CRDT claimed advantages over OT. We hope the results reported in this paper will help clear up common myths, misconceptions, and confusions surrounding alternative co-editing techniques, and accelerate progress in co-editing technology for real world applications.

cs.DC↗

arXiv2011

Thermal Properties of Isotopically Engineered Graphene

In addition to its exotic electronic properties graphene exhibits unusually high intrinsic thermal conductivity. The physics of phonons - the main heat carriers in graphene - was shown to be substantially different in two-dimensional (2D) crystals, such as graphene, than in three-dimensional (3D) graphite. Here, we report our experimental study of the isotope effects on the thermal properties of graphene. Isotopically modified graphene containing various percentages of 13C were synthesized by chemical vapor deposition (CVD). The regions of different isotopic composition were parts of the same graphene sheet to ensure uniformity in material parameters. The thermal conductivity, K, of isotopically pure 12C (0.01% 13C) graphene determined by the optothermal Raman technique, was higher than 4000 W/mK at the measured temperature Tm~320 K, and more than a factor of two higher than the value of K in a graphene sheets composed of a 50%-50% mixture of 12C and 13C. The experimental data agree well with our molecular dynamics (MD) simulations, corrected for the long-wavelength phonon contributions via the Klemens model. The experimental results are expected to stimulate further studies aimed at better understanding of thermal phenomena in 2D crystals.

cond-mat.mes-hall↗

arXiv2010

Oxidation resistance of graphene-coated Cu and Cu/Ni alloy

The ability to protect refined metals from reactive environments is vital to many industrial and academic applications. Current solutions, however, typically introduce several negative effects, including increased thickness and changes in the metal physical properties. In this paper, we demonstrate for the first time the ability of graphene films grown by chemical vapor deposition to protect the surface of the metallic growth substrates of Cu and Cu/Ni alloy from air oxidation. SEM, Raman spectroscopy, and XPS studies show that the metal surface is well protected from oxidation even after heating at 200 \degree C in air for up to 4 hours. Our work further shows that graphene provides effective resistance against hydrogen peroxide. This protection method offers significant advantages and can be used on any metal that catalyzes graphene growth.

cond-mat.mtrl-sci↗

arXiv2010

Adsorption/desorption and electrically controlled flipping of ammonia molecules on graphene

In this paper, we evaluate of the adsorption/ desorption of ammonia molecules on a graphene surface by studying the Fermi level shift. Based on a physically plausible model, the adsorption and desorption rates of ammonia molecules on graphene have been extracted from the measured Fermi level shift as a function of exposure time. An electric field-induced flipping behavior of ammonia molecules on graphene is suggested, based on field effect transistor (FET) measurements.

cond-mat.mes-hall↗

arXiv2010

Graphene films with large domain size by a two-step chemical vapor deposition process

The fundamental properties of graphene are making it an attractive material for a wide variety of applications. Various techniques have been developed to produce graphene and recently we discovered the synthesis of large area graphene by chemical vapor deposition (CVD) of methane on Cu foils. We also showed that graphene growth on Cu is a surface-mediated process and the films were polycrystalline with domains having an area of tens of square microns. In this paper we report on the effect of growth parameters such as temperature, and methane flow rate and partial pressure on the growth rate, domain size, and surface coverage of graphene as determined by Raman spectroscopy, and transmission and scanning electron microscopy. Based on the results, we developed a two-step CVD process to synthesize graphene films with domains having an area of hundreds of square microns. Scanning electron microscopy and Raman spectroscopy clearly show an increase in domain size by changing the growth parameters. Transmission electron microscopy further shows that the domains are crystallographically rotated with respect to each other with a range of angles from about 13 degrees to nearly 30 degrees. Electrical transport measurements performed on back-gated FETs show that overall films with larger domains tend to have higher carrier mobility, up to about 16,000 cm2 V-1 s-1 at room temperature.

cond-mat.mtrl-sci↗

arXiv2009

Evolution of graphene growth on Cu and Ni studied by carbon isotope labeling

Large-area graphene is a new material with properties that make it desirable for advanced scaled electronic devices1. Recently, chemical vapor deposition (CVD) of graphene and few-layer graphene using hydrocarbons on metal substrates such as Ni and Cu has shown to be a promising technique2-5. It has been proposed in recent publications that graphene growth on Ni occurs by C segregation2 or precipitation3, while that on Cu is by surface adsorption5. In this letter, we used a carbon isotope labeling technique to elucidate the growth kinetics and unambiguously demonstrate that graphene growth on Cu is by surface adsorption whereas on Ni is by segregation-precipitation. An understanding of the evolution of graphene growth and thus growth mechanism(s) is desired to obtain uniform graphene films. The results presented in this letter clearly demonstrate that surface adsorption is preferred over precipitation to grow graphene because it is a self-limiting process and thus manufacturable.

cond-mat.mtrl-sci↗

arXiv2009

Large-Area Synthesis of High-Quality and Uniform Graphene Films on Copper Foils

Graphene has been attracting great interest because of its distinctive band structure and physical properties. Today, graphene is limited to small sizes because it is produced mostly by exfoliating graphite. We grew large-area graphene films of the order of centimeters on copper substrates by chemical vapor deposition using methane. The films are predominantly single layer graphene with a small percentage (less than 5%) of the area having few layers, and are continuous across copper surface steps and grain boundaries. The low solubility of carbon in copper appears to help make this growth process self-limiting. We also developed graphene film transfer processes to arbitrary substrates, and dual-gated field-effect transistors fabricated on Si/SiO2 substrates showed electron mobilities as high as 4050 cm2V-1s-1 at room temperature.

cond-mat.mtrl-sci↗

arXiv2007

Photoconductivity of Single-crystalline Selenium Nanotubes

Photoconductivity of single-crystalline selenium nanotubes (SCSNT) under a range of illumination intensities of a 633nm laser is carried out with a novel two terminal device arrangement at room temperature. It's found that SCSNT forms Schottky barriers with the W and Au contacts, and the barrier height is a function of the light intensities. In low illumination regime below 1.46x10E-4 muWmum-2, the Au-Se-W hybrid structure exhibits sharp switch on/off behavior, and the turn-on voltages decrease with increasing illuminating intensities. In the high illumination regime above 7x10E-4 muWmum-2, the device exhibits ohmic conductance with a photoconductivity as high as 0.59Ohmcm-1, significantly higher that reported values for carbon and GaN nanotubes. This finding suggests that SCSNT is potentially a good photo-sensor material as well we a very effective solar cell material.

cond-mat.mtrl-sci↗