BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Guanxiong Liu

Guanxiong Liu appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
cond-mat.mtrl-scicond-mat.mes-hallcond-mat.otherCryptography and SecurityMachine LearningEmerging Technologies

Catalog footprint

What is connected

15works
6topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

15 published item(s)

preprint2020arXiv

ManiGen: A Manifold Aided Black-box Generator of Adversarial Examples

Machine learning models, especially neural network (NN) classifiers, have acceptable performance and accuracy that leads to their wide adoption in different aspects of our daily lives. The underlying assumption is that these models are generated and used in attack free scenarios. However, it has been shown that neural network based classifiers are vulnerable to adversarial examples. Adversarial examples are inputs with special perturbations that are ignored by human eyes while can mislead NN classifiers. Most of the existing methods for generating such perturbations require a certain level of knowledge about the target classifier, which makes them not very practical. For example, some generators require knowledge of pre-softmax logits while others utilize prediction scores. In this paper, we design a practical black-box adversarial example generator, dubbed ManiGen. ManiGen does not require any knowledge of the inner state of the target classifier. It generates adversarial examples by searching along the manifold, which is a concise representation of input data. Through extensive set of experiments on different datasets, we show that (1) adversarial examples generated by ManiGen can mislead standalone classifiers by being as successful as the state-of-the-art white-box generator, Carlini, and (2) adversarial examples generated by ManiGen can more effectively attack classifiers with state-of-the-art defenses.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Using Single-Step Adversarial Training to Defend Iterative Adversarial Examples

Adversarial examples have become one of the largest challenges that machine learning models, especially neural network classifiers, face. These adversarial examples break the assumption of attack-free scenario and fool state-of-the-art (SOTA) classifiers with insignificant perturbations to human. So far, researchers achieved great progress in utilizing adversarial training as a defense. However, the overwhelming computational cost degrades its applicability and little has been done to overcome this issue. Single-Step adversarial training methods have been proposed as computationally viable solutions, however they still fail to defend against iterative adversarial examples. In this work, we first experimentally analyze several different SOTA defense methods against adversarial examples. Then, based on observations from experiments, we propose a novel single-step adversarial training method which can defend against both single-step and iterative adversarial examples. Lastly, through extensive evaluations, we demonstrate that our proposed method outperforms the SOTA single-step and iterative adversarial training defense. Compared with ATDA (single-step method) on CIFAR10 dataset, our proposed method achieves 35.67% enhancement in test accuracy and 19.14% reduction in training time. When compared with methods that use BIM or Madry examples (iterative methods) on CIFAR10 dataset, it saves up to 76.03% in training time with less than 3.78% degeneration in test accuracy.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

An Integrated Tantalum Sulfide - Boron Nitride - Graphene Oscillator: A Charge-Density-Wave Device Operating at Room Temperature

The charge-density-wave (CDW) phase is a macroscopic quantum state consisting of a periodic modulation of the electronic charge density accompanied by a periodic distortion of the atomic lattice in quasi-1D or layered 2D metallic crystals. Several layered transition metal dichalcogenides, such as 1T-TaSe2, 1T-TaS2 and 1T-TiSe2, exhibit unusually high transition temperatures to different CDW symmetry-reducing phases. These transitions can be affected by environmental conditions, film thickness and applied electric bias. However, device applications of these intriguing systems at room temperature or their integration with other 2D materials have not been explored. Here we show that in 2D CDW 1T-TaS2, the abrupt change in the electrical conductivity and hysteresis at the transition point between nearly-commensurate and incommensurate charge-density-wave phases can be used for constructing an oscillator that operates at room temperature. The hexagonal boron nitride was capped on 1T-TaS2 thin film to provide protection from oxidation, and an integrated graphene transistor provides a voltage tunable, matched, low-resistance load enabling precise voltage control of the oscillator frequency. The integration of these three disparate two-dimensional materials, in a way that exploits the unique properties of each, yields a simple, miniaturized, voltage-controlled oscillator device. Theoretical considerations suggest that the upper limit of oscillation frequency to be in the THz regime.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Two-Dimensional Oscillatory Neural Network Based on Charge-Density-Wave Devices Operating at Room Temperature

We propose an oscillatory neural network implemented with two-dimensional tantalum disulfide devices operating in the change density wave regime at room temperature. An elementary cell of the network consists of two 1T-TaS2 devices connected in series. Such a cell has constant output and oscillatory states. All cells have the same bias voltage. There is constant current flowing through the cell in the constant output mode. The oscillations occur at a certain bias voltage due to the electrical-field driven metal-to-insulator transition owing to the changes in the charge density wave phase in the 1T-TaS2 channel. Two 1T-TaS2 devices oscillate out-of-phase where one of the devices is in the insulator phase while the other one is in the metallic state. The nearest-neighbor cells are coupled via graphene transistors. The cells are resistively coupled if the graphene transistor is in the On state while they are capacitively coupled if the transistor is in the Off state. The operation of the oscillatory neural network is simulated numerically for the 30x30 node network. The results of our numerical modeling show the formation of artificial vortexes and cellular-automata type data processing. The two-dimensional 1T-TaS2 devices, utilized in the network, offer a unique combination of properties such as scalability, high operational frequency, fast synchronization speed, and radiation hardness, which makes them promising for both consumer electronic and defense applications.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Suppression of 1/f Noise in Near-Ballistic h-BN-Graphene-h-BN Heterostructure Field-Effect Transistors

We have investigated low-frequency 1/f noise in the boron nitride - grapheme - boron nitride heterostructure field - effect transistors on Si/SiO2 substrates (f is a frequency). The device channel was implemented with a single layer graphene encased between two layers of hexagonal boron nitride. The transistors had the charge carrier mobility in the range from 30000 to 36000 cm2/Vs at room temperature. It was established that the noise spectral density normalized to the channel area in such devices can be suppressed to 5 x 10^-9 μm2 Hz^-1, which is a factor of x5 - x10 lower than that in non-encapsulated graphene devices on Si/SiO2. The physical mechanism of noise suppression was attributed to screening of the charge carriers in the channel from traps in SiO2 gate dielectric and surface defects. The obtained results are important for the electronic and optoelectronic applications of graphene.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Graphene-Based Non-Boolean Logic Circuits

Graphene revealed a number of unique properties beneficial for electronics. However, graphene does not have an energy band-gap, which presents a serious hurdle for its applications in digital logic gates. The efforts to induce a band-gap in graphene via quantum confinement or surface functionalization have not resulted in a breakthrough. Here we show that the negative differential resistance experimentally observed in graphene field-effect transistors of "conventional" design allows for construction of viable non-Boolean computational architectures with the gap-less graphene. The negative differential resistance - observed under certain biasing schemes - is an intrinsic property of graphene resulting from its symmetric band structure. Our atomistic modeling shows that the negative differential resistance appears not only in the drift-diffusion regime but also in the ballistic regime at the nanometer-scale - although the physics changes. The obtained results present a conceptual change in graphene research and indicate an alternative route for graphene's applications in information processing.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

Direct Probing of 1/f Noise Origin with Graphene Multilayers: Surface vs. Volume

Low-frequency noise with the spectral density S(f)~1/f^g (f is the frequency and g~1) is a ubiquitous phenomenon, which hampers operation of many devices and circuits. A long-standing question of particular importance for electronics is whether 1/f noise is generated on the surface of electrical conductors or inside their volumes. Using high-quality graphene multilayers we were able to directly address this fundamental problem of the noise origin. Unlike the thickness of metal or semiconductor films, the thickness of graphene multilayers can be continuously and uniformly varied all the way down to a single atomic layer of graphene - the actual surface. We found that 1/f noise becomes dominated by the volume noise when the thickness exceeds ~7 atomic layers (~2.5 nm). The 1/f noise is the surface phenomenon below this thickness. The obtained results are important for continuous downscaling of conventional electronics and for the proposed graphene applications in sensors and communications.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

Graphene-Graphite Quilts for Thermal Management of High-Power GaN Transistors

Self-heating is a severe problem for high-power GaN electronic and optoelectronic devices. Various thermal management solutions, e.g. flip-chip bonding or composite substrates have been attempted. However, temperature rise still limits applications of the nitride-based technology. Here we demonstrate that thermal management of GaN transistors can be substantially improved via introduction of the alternative heat-escaping channels implemented with few-layer graphene - an excellent heat conductor. We have transferred few-layer graphene to AlGaN/GaN heterostructure field-effect transistors on SiC substrates to form the "graphene-graphite quilts" - lateral heat spreaders, which remove heat from the channel regions. Using the micro-Raman spectroscopy for in-situ monitoring we have shown that temperature can be lowered by as much as ~ 20oC in such devices operating at ~13-W/mm power density. The simulations suggest that the efficiency of the "graphene quilts" can be made even higher in GaN devices on thermally resistive sapphire substrates and in the designs with the closely located heat sinks. Our results open a novel application niche for few-layer graphene in high-power electronics.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

Graphene-on-Diamond Devices with Enhanced Current-Carrying Capacity: Carbon sp2-on-sp3 Technology

Graphene demonstrated potential for practical applications owing to its excellent electronic and thermal properties. Typical graphene field-effect transistors and interconnects built on conventional SiO2/Si substrates reveal the breakdown current density on the order of 1 uA/nm2 (i.e. 10^8 A/cm2) which is ~100\times larger than the fundamental limit for the metals but still smaller than the maximum achieved in carbon nanotubes. We show that by replacing SiO2 with synthetic diamond one can substantially increase the current-carrying capacity of graphene to as high as ~18 uA/nm2 even at ambient conditions. Our results indicate that graphene's current-induced breakdown is thermally activated. We also found that the current carrying capacity of graphene can be improved not only on the single-crystal diamond substrates but also on an inexpensive ultrananocrystalline diamond, which can be produced in a process compatible with a conventional Si technology. The latter was attributed to the decreased thermal resistance of the ultrananocrystalline diamond layer at elevated temperatures. The obtained results are important for graphene's applications in high-frequency transistors, interconnects, transparent electrodes and can lead to the new planar sp2-on-sp3 carbon-on-carbon technology.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

Selective Gas Sensing with a Single Pristine Graphene Transistor

We show that vapors of different chemicals produce distinguishably different effects on the low-frequency noise spectra of graphene. It was found in a systematic study that some gases change the electrical resistance of graphene devices without changing their low-frequency noise spectra while other gases modify the noise spectra by inducing Lorentzian components with distinctive features. The characteristic frequency fc of the Lorentzian noise bulges in graphene devices is different for different chemicals and varies from fc=10 - 20 Hz to fc=1300 - 1600 Hz for tetrahydrofuran and chloroform vapors, respectively. The obtained results indicate that the low-frequency noise in combination with other sensing parameters can allow one to achieve the selective gas sensing with a single pristine graphene transistor. Our method of gas sensing with graphene does not require graphene surface functionalization or fabrication of an array of the devices with each tuned to a certain chemical.

0 citations0 reviewsNo code linkedOpen access
preprint2011arXiv

Graphene Thickness-Graded Transistors with Reduced Low-Frequency 1/f Noise

We demonstrate graphene thickness-graded transistors with high electron mobility and low 1/f noise (f is a frequency). The device channel is implemented with few-layer graphene with the thickness varied from a single layer in the middle to few-layers at the source and drain contacts. It was found that such devices have electron mobility comparable to the reference single-layer graphene devices while producing lower noise levels. The metal doping of graphene and difference in the electron density of states between the single-layer and few-layer graphene cause the observed noise reduction. The results shed light on the noise origin in graphene.

0 citations0 reviewsNo code linkedOpen access
preprint2011arXiv

Observation of the "Memory Steps" in Graphene at Elevated Temperatures

We found that the current-voltage characteristics of the single-layer graphene field-effect transistors exhibit an intriguing feature - an abrupt change of the current near zero gate bias at elevated temperatures T > 500 K. The strength of the effect - referred to as the "memory step" by analogy with the "memory dips" - known phenomenon in electron glasses - depends on the rate of the voltage sweep. The slower the sweep - the more pronounced is the step in the current. Despite differences in examined graphene transistor characteristics, the "memory step" always appears near zero gate bias. The effect is reproducible and preserved after device aging. A similar feature has been previously observed in electronic glasses albeit at cryogenic temperatures and with opposite dependence on the rate of the voltage sweep. The observed "memory step" can be related to the slow relaxation processes in graphene. This new characteristic of electron transport in graphene can be used for applications in high-temperature sensors and switches.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Growth of Large-Area Graphene Films from Metal-Carbon Melts

We have demonstrated a new method for the large-area graphene growth, which can lead to a scalable low-cost high-throughput production technology. The method is based on growing single-layer or few-layer graphene films from a molten phase. The process involves dissolving carbon inside a molten metal at a specified temperature and then allowing the dissolved carbon to nucleate and grow on top of the melt at a lower temperature. The examined metals for the metal - carbon melts included copper and nickel. For the latter, pristine single layer graphene was grown successfully. The resulting graphene layers were subjected to detailed microscopic and Raman spectroscopic characterization. The deconvolution of the Raman 2D band was used to accurately determine the number of atomic planes in the resulting graphene layers and access their quality. The results indicate that our technology can provide bulk graphite films, few-layer graphene as well as high-quality single layer graphene on metals. Our approach can also be used for producing graphene-metal thermal interface materials for thermal management applications.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Triple-Mode Single-Transistor Graphene Amplifier and Its Applications

In this article, we propose and experimentally demonstrate a triple-mode single-transistor graphene amplifier utilizing a three-terminal back-gated single-layer graphene transistor. The ambipolar nature of electronic transport in graphene transistors leads to increased amplifier functionality as compared to amplifiers built with unipolar semiconductor devices. The ambipolar graphene transistors can be configured as n-type, p-type, or hybrid-type by changing the gate bias. As a result, the single-transistor graphene amplifier can operate in the common-source, common-drain, or frequency multiplication mode, respectively. This in-field controllability of the single-transistor graphene amplifier can be used to realize the modulation necessary for phase shift keying and frequency shift keying, which are widely used in wireless applications. It also offers new opportunities for designing analog circuits with simpler structure and higher integration densities for communications applications.

0 citations0 reviewsNo code linkedOpen access
preprint2009arXiv

Ultraviolet Raman Spectroscopy of Single and Multi-layer Graphene

We investigated Raman spectra of single-layer and multi-layer graphene under ultraviolet laser excitation at the wavelength of 325 nm. It was found that while the G peak of graphene remains pronounced in UV Raman spectra, the 2D band intensity undergoes severe quenching. The evolution of the ratio of the intensities of the G and 2D peaks, I(G)/I(2D), as the number of graphene layers n changes from n=1 to n=5, is different in UV Raman spectra from that in conventional visible Raman spectra excited at the 488 nm and 633 nm wavelengths. The 2D band under UV excitation shifts to larger wave numbers and is found near 2825 1/cm. The observed UV Raman features of graphene were explained by invoking the resonant scattering model. The obtained results contribute to the Raman nanometrology of graphene by providing an additional metric for determining the number of graphene layers and assessing its quality.

0 citations0 reviewsNo code linkedOpen access