arXiv subjects

Haiyan Wang

Publications and source records attributed to Haiyan Wang.

At least 73 records · Page 4Linked to original sources

arXiv2022

A Nanometer-Thick Oxide Semiconductor Transistor with Ultra-High Drain Current

High drive current is a critical performance parameter in semiconductor devices for high-speed, low-power logic applications or high-efficiency, high-power, high-speed radio frequency (RF) analog applications. In this work, we demonstrate an In2O3 transistor grown by atomic layer deposition (ALD) at back-end-of-line (BEOL) compatible temperatures with a record high drain current exceeding 10 A/mm, the performance of which is 2-3 times better than all known transistors with semiconductor channels. A record high transconductance of 4 S/mm is also achieved among all transistors with a planar structure. It is found that a high carrier density and high electron velocity both contribute to this remarkably high on-state performance in ALD In2O3 transistors, which is made possible by the high-quality oxide/oxide interface, the metal-like charge-neutrality-level (CNL) alignment, and the high band velocities induced by the low density-of-state (DOS). Experimental Hall, I-V and split C-V measurements at room temperature confirm a high carrier density up to 6-7*10^13 /cm2 and a high velocity of about 10^7 cm/s. Ultra-thin oxide semiconductors, with a CNL located deep inside the conduction band, represent a promising new direction for the search of alternative channel materials for high-performance semiconductor devices.

cond-mat.mes-hall↗

arXiv2022

A Gate-All-Around Single-Channel In2O3 Nanoribbon FET with Near 20 mA/μm Drain Current

In this work, we demonstrate atomic-layer-deposited (ALD) single-channel indium oxide (In2O3) gate-all-around (GAA) nanoribbon FETs in a back-end-of-line (BEOL) compatible process. A maximum on-state current (ION) of 19.3 mA/μm (near 20 mA/μm) is achieved in an In2O3 GAA nanoribbon FET with a channel thickness (TIO) of 3.1 nm, channel length (Lch) of 40 nm, channel width (Wch) of 30 nm and dielectric HfO2 of 5 nm. The record high drain current obtained from an In2O3 FET is about one order of magnitude higher than any conventional single-channel semiconductor FETs. This extraordinary drain current and its related on-state performance demonstrate ALD In2O3 is a promising oxide semiconductor channel with great opportunities in BEOL compatible monolithic 3D integration.

physics.app-ph↗

arXiv2022

Sequential Point Clouds: A Survey

Point cloud has drawn more and more research attention as well as real-world applications. However, many of these applications (e.g. autonomous driving and robotic manipulation) are actually based on sequential point clouds (i.e. four dimensions) because the information of the static point cloud data could provide is still limited. Recently, researchers put more and more effort into sequential point clouds. This paper presents an extensive review of the deep learning-based methods for sequential point cloud research including dynamic flow estimation, object detection \& tracking, point cloud segmentation, and point cloud forecasting. This paper further summarizes and compares the quantitative results of the reviewed methods over the public benchmark datasets. Finally, this paper is concluded by discussing the challenges in the current sequential point cloud research and pointing out insightful potential future research directions.

cs.CV↗

arXiv2022

PSMNet: Position-aware Stereo Merging Network for Room Layout Estimation

In this paper, we propose a new deep learning-based method for estimating room layout given a pair of 360 panoramas. Our system, called Position-aware Stereo Merging Network or PSMNet, is an end-to-end joint layout-pose estimator. PSMNet consists of a Stereo Pano Pose (SP2) transformer and a novel Cross-Perspective Projection (CP2) layer. The stereo-view SP2 transformer is used to implicitly infer correspondences between views, and can handle noisy poses. The pose-aware CP2 layer is designed to render features from the adjacent view to the anchor (reference) view, in order to perform view fusion and estimate the visible layout. Our experiments and analysis validate our method, which significantly outperforms the state-of-the-art layout estimators, especially for large and complex room spaces.

cs.CV↗

arXiv2022

Experimental Quantum End-to-End Learning on a Superconducting Processor

Machine learning can be substantially powered by a quantum computer owing to its huge Hilbert space and inherent quantum parallelism. In the pursuit of quantum advantages for machine learning with noisy intermediate-scale quantum devices, it was proposed that the learning model can be designed in an end-to-end fashion, i.e., the quantum ansatz is parameterized by directly manipulable control pulses without circuit design and compilation. Such gate-free models are hardware friendly and can fully exploit limited quantum resources. Here, we report the first experimental realization of quantum end-to-end machine learning on a superconducting processor. The trained model can achieve 98% recognition accuracy for two handwritten digits (via two qubits) and 89% for four digits (via three qubits) in the MNIST (Mixed National Institute of Standards and Technology) database. The experimental results exhibit the great potential of quantum end-to-end learning for resolving complex real-world tasks when more qubits are available.

quant-ph↗

arXiv2022

Scaled indium oxide transistors fabricated using atomic layer deposition

In order to continue to improve integrated circuit performance and functionality, scaled transistors with short channel lengths and low thickness are needed. But the further scaling of silicon-based devices and the development of alternative semiconductor channel materials that are compatible with current fabrication processes is challenging. Here we report atomic-layer-deposited indium oxide transistors with channel lengths down to 8 nm, channel thicknesses down to 0.5 nm and equivalent dielectric oxide thickness down to 0.84 nm. Due to the scaled device dimensions and low contact resistance, the devices exhibit high on-state currents of 3.1 A/mm at a drain voltage of 0.5 V and a transconductance of 1.5 S/mm at a drain voltage 1 V. Our devices are a promising alternative channel material for scaled transistors with back-end-of-line processing compatibility.

cond-mat.mtrl-sci↗

arXiv2022

K-nearest Multi-agent Deep Reinforcement Learning for Collaborative Tasks with a Variable Number of Agents

Traditionally, the performance of multi-agent deep reinforcement learning algorithms are demonstrated and validated in gaming environments where we often have a fixed number of agents. In many industrial applications, the number of available agents can change at any given day and even when the number of agents is known ahead of time, it is common for an agent to break during the operation and become unavailable for a period of time. In this paper, we propose a new deep reinforcement learning algorithm for multi-agent collaborative tasks with a variable number of agents. We demonstrate the application of our algorithm using a fleet management simulator developed by Hitachi to generate realistic scenarios in a production site.

cs.LG↗

arXiv2021

A generalized 3D elastic model for nanoscale, self-assembled oxide-metal thin films with pillar-in-matrix configurations

In recent years, functional oxide-metal based vertically aligned nanocomposite (VAN) thin films have gained interest due to their intriguing physical properties and multifunctionalities stemming from the complex interactions between the two phases in the film and the substrate. In this work, we develop a model for studying the energetics of these thin film systems, including the effects of both lattice mismatch and capillary forces due to interface curvature. Each phase is incorporated into the model using a phase indicator function, and we introduce the capillary forces as body forces using a vector density representation of the interface. The model is implemented using the finite element method to study the deformation of the thin film which is composed of Au nanopillars embedded in a La0.7Sr0.3MnO3 (LSMO) matrix on an SrTiO3 (STO) substrate. The results suggest that the total energy is lowest for random configurations of pillars compared to ordered square and hexagonal lattice configurations, consistent with the random distribution of pillars found in experiments. Furthermore, we find that the interfacial energy dominates the total energy of each configuration, suggesting that interfacial energy in the system is an important design parameter for nanocomposite growth, along with the lattice mismatch.

cond-mat.mtrl-sci↗

arXiv2021

FESTA: Flow Estimation via Spatial-Temporal Attention for Scene Point Clouds

Scene flow depicts the dynamics of a 3D scene, which is critical for various applications such as autonomous driving, robot navigation, AR/VR, etc. Conventionally, scene flow is estimated from dense/regular RGB video frames. With the development of depth-sensing technologies, precise 3D measurements are available via point clouds which have sparked new research in 3D scene flow. Nevertheless, it remains challenging to extract scene flow from point clouds due to the sparsity and irregularity in typical point cloud sampling patterns. One major issue related to irregular sampling is identified as the randomness during point set abstraction/feature extraction -- an elementary process in many flow estimation scenarios. A novel Spatial Abstraction with Attention (SA^2) layer is accordingly proposed to alleviate the unstable abstraction problem. Moreover, a Temporal Abstraction with Attention (TA^2) layer is proposed to rectify attention in temporal domain, leading to benefits with motions scaled in a larger range. Extensive analysis and experiments verified the motivation and significant performance gains of our method, dubbed as Flow Estimation via Spatial-Temporal Attention (FESTA), when compared to several state-of-the-art benchmarks of scene flow estimation.

cs.CV↗

arXiv2021

Deep Reinforcement Learning with Adjustments

Deep reinforcement learning (RL) algorithms can learn complex policies to optimize agent operation over time. RL algorithms have shown promising results in solving complicated problems in recent years. However, their application on real-world physical systems remains limited. Despite the advancements in RL algorithms, the industries often prefer traditional control strategies. Traditional methods are simple, computationally efficient and easy to adjust. In this paper, we first propose a new Q-learning algorithm for continuous action space, which can bridge the control and RL algorithms and bring us the best of both worlds. Our method can learn complex policies to achieve long-term goals and at the same time it can be easily adjusted to address short-term requirements without retraining. Next, we present an approximation of our algorithm which can be applied to address short-term requirements of any pre-trained RL algorithm. The case studies demonstrate that both our proposed method as well as its practical approximation can achieve short-term and long-term goals without complex reward functions.

cs.LG↗

arXiv2021

An Offline Deep Reinforcement Learning for Maintenance Decision-Making

Several machine learning and deep learning frameworks have been proposed to solve remaining useful life estimation and failure prediction problems in recent years. Having access to the remaining useful life estimation or likelihood of failure in near future helps operators to assess the operating conditions and, therefore, provides better opportunities for sound repair and maintenance decisions. However, many operators believe remaining useful life estimation and failure prediction solutions are incomplete answers to the maintenance challenge. They argue that knowing the likelihood of failure in the future is not enough to make maintenance decisions that minimize costs and keep the operators safe. In this paper, we present a maintenance framework based on offline supervised deep reinforcement learning that instead of providing information such as likelihood of failure, suggests actions such as "continuation of the operation" or "the visitation of the repair shop" to the operators in order to maximize the overall profit. Using offline reinforcement learning makes it possible to learn the optimum maintenance policy from historical data without relying on expensive simulators. We demonstrate the application of our solution in a case study using the NASA C-MAPSS dataset.

cs.LG↗

arXiv2021

Intrinsic ferroelectricity in Y-doped HfO2 thin films

Ferroelectric HfO2-based materials hold great potential for widespread integration of ferroelectricity into modern electronics due to their robust ferroelectric properties at the nanoscale and compatibility with the existing Si technology. Earlier work indicated that the nanometer crystal grain size was crucial for stabilization of the ferroelectric phase of hafnia. This constraint caused high density of unavoidable structural defects of the HfO2-based ferroelectrics, obscuring the intrinsic ferroelectricity inherited from the crystal space group of bulk HfO2. Here, we demonstrate the intrinsic ferroelectricity in Y-doped HfO2 films of high crystallinity. Contrary to the common expectation, we show that in the 5% Y-doped HfO2 epitaxial thin films, high crystallinity enhances the spontaneous polarization up to a record-high 50 μC/cm2 value at room temperature. The high spontaneous polarization persists at reduced temperature, with polarization values consistent with our theoretical predictions, indicating the dominant contribution from the intrinsic ferroelectricity. The crystal structure of these films reveals the Pca21 orthorhombic phase with a small rhombohedral distortion, underlining the role of the anisotropic stress and strain. These results open a pathway to controlling the intrinsic ferroelectricity in the HfO2-based materials and optimizing their performance in applications.

cond-mat.mtrl-sci↗

arXiv2021

High-efficiency arbitrary quantum operation on a high-dimensional quantum system

The ability to manipulate quantum systems lies at the heart of the development of quantum technology. The ultimate goal of quantum control is to realize arbitrary quantum operations (AQuOs) for all possible open quantum system dynamics. However, the demanding extra physical resources impose great obstacles. Here, we experimentally demonstrate a universal approach of AQuO on a photonic qudit with minimum physical resource of a two-level ancilla and a $\log_{2}d$-scale circuit depth for a $d$-dimensional system. The AQuO is then applied in quantum trajectory simulation for quantum subspace stabilization and quantum Zeno dynamics, as well as incoherent manipulation and generalized measurements of the qudit. Therefore, the demonstrated AQuO for complete quantum control would play an indispensable role in quantum information science.

quant-ph↗

arXiv2021

Experimental simulation of open quantum system dynamics via Trotterization

Digital quantum simulators provide a diversified tool for solving the evolution of quantum systems with complicated Hamiltonians and hold great potential for a wide range of applications. Although much attention is paid to the unitary evolution of closed quantum systems, dissipation and noise are vital in understanding the dynamics of practical quantum systems. In this work, we experimentally demonstrate a digital simulation of an open quantum system in a controllable Markovian environment with the assistance of a single ancillary qubit. By Trotterizing the quantum Liouvillians, the continuous evolution of an open quantum system is effectively realized, and its application in error mitigation is demonstrated by adjusting the simulated noise intensities. High-order Trotter for open quantum dynamics is also experimentally investigated and shows higher accuracy. Our results represent a significant step towards hardware-efficient simulation of open quantum systems and error mitigation in quantum algorithms in noisy intermediate-scale quantum systems.

quant-ph↗

arXiv2021

Route to High-Performance Micro-solid Oxide Fuel Cells on Metallic Substrates

Micro-solid oxide fuel cells based on thin films have strong potential for use in portable power devices. However, devices based on silicon substrates typically involve thin-film metallic electrodes which are unstable at high temperatures. Devices based on bulk metal substrates overcome these limitations, though performance is hindered by the challenge of growing state-of-the-art epitaxial materials on metals. Here, we demonstrate for the first time the growth of epitaxial cathode materials on metal substrates (stainless steel) commercially supplied with epitaxial electrolyte layers (1.5 {um (Y2O3)0.15(ZrO2)0.85 (YSZ) + 50 nm CeO2). We create epitaxial mesoporous cathodes of (La0.60Sr0.40)0.95Co0.20Fe0.80O3 (LSCF) on the substrate by growing LSCF/MgO vertically aligned nanocomposite films by pulsed laser deposition, followed by selectively etching out the MgO. To enable valid comparison with the literature, the cathodes are also grown on single-crystal substrates, confirming state-of-the-art performance with an area specific resistance of 100ohmegacm2 at 500dC and activation energy down to 0.97 eV. The work marks an important step toward the commercialization of high-performance micro-solid oxide fuel cells for portable power applications.

physics.app-ph↗

arXiv2020

Interactions of Linguistic and Domain Overhypotheses in Category Learning

For humans learning to categorize and distinguish parts of the world, the set of assumptions (overhypotheses) they hold about potential category structures is directly related to their learning process. In this work we examine the effects of two overhypotheses for category learning: 1) the bias introduced by the presence of linguistic labels for objects; 2) the conceptual 'domain' biases inherent in the learner about which features are most indicative of category structure. These two biases work in tandem to impose priors on the learning process; and we model and detail their interaction and effects. This paper entails an adaptation and expansion of prior experimental work that addressed label bias effects but did not fully explore conceptual domain biases. Our results highlight the importance of both the domain and label biases in facilitating or hindering category learning.

cs.CE↗

arXiv2020

Why In2O3 Can Make 0.7 nm Atomic Layer Thin Transistors?

In this work, we demonstrate enhancement-mode field-effect transistors by atomic-layer-deposited (ALD) amorphous In2O3 channel with thickness down to 0.7 nm. Thickness is found to be critical on the materials and electron transport of In2O3. Controllable thickness of In2O3 at atomic scale enables the design of sufficient 2D carrier density in the In2O3 channel integrated with the conventional dielectric. The threshold voltage and channel carrier density are found to be considerably tuned by channel thickness. Such phenomenon is understood by the trap neutral level (TNL) model where the Fermi-level tends to align deeply inside the conduction band of In2O3 and can be modulated to the bandgap in atomic layer thin In2O3 due to quantum confinement effect, which is confirmed by density function theory (DFT) calculation. The demonstration of enhancement-mode amorphous In2O3 transistors suggests In2O3 is a competitive channel material for back-end-of-line (BEOL) compatible transistors and monolithic 3D integration applications.

cond-mat.mtrl-sci↗

arXiv2020

A Non-linear Function-on-Function Model for Regression with Time Series Data

In the last few decades, building regression models for non-scalar variables, including time series, text, image, and video, has attracted increasing interests of researchers from the data analytic community. In this paper, we focus on a multivariate time series regression problem. Specifically, we aim to learn mathematical mappings from multiple chronologically measured numerical variables within a certain time interval S to multiple numerical variables of interest over time interval T. Prior arts, including the multivariate regression model, the Seq2Seq model, and the functional linear models, suffer from several limitations. The first two types of models can only handle regularly observed time series. Besides, the conventional multivariate regression models tend to be biased and inefficient, as they are incapable of encoding the temporal dependencies among observations from the same time series. The sequential learning models explicitly use the same set of parameters along time, which has negative impacts on accuracy. The function-on-function linear model in functional data analysis (a branch of statistics) is insufficient to capture complex correlations among the considered time series and suffer from underfitting easily. In this paper, we propose a general functional mapping that embraces the function-on-function linear model as a special case. We then propose a non-linear function-on-function model using the fully connected neural network to learn the mapping from data, which addresses the aforementioned concerns in the existing approaches. For the proposed model, we describe in detail the corresponding numerical implementation procedures. The effectiveness of the proposed model is demonstrated through the application to two real-world problems.

cs.LG↗