arXiv subjects

Feng Gao

Publications and source records attributed to Feng Gao.

At least 271 records · Page 15Linked to original sources

arXiv2020

Limited-view Photoacoustic Imaging Reconstruction With Dual Domain Inputs Under Mutual Information Constraint

Based on photoacoustic effect, photoacoustic tomography is developing very fast in recent years, and becoming an important imaging tool for both preclinical and clinical studies. With enough ultrasound transducers placed around the biological tissue, PAT can provide both deep penetration and high image contrast by hybrid usage of light and sound. However, considering space and measurement environmental limitations, transducers are always placed in a limited-angle way, which means that the other side without transducer coverage suffers severe information loss. With conventional image reconstruction algorithms, the limited-view tissue induces artifacts and information loss, which may cause doctors misdiagnosis or missed diagnosis. In order to solve limited-view PA imaging reconstruction problem, we propose to use both time domain and frequency domain reconstruction algorithms to get delay-and-sum (DAS) image inputs and k-space image inputs. These dual domain images share nearly same texture information but different artifact information, which can teach network how to distinguish these two kinds of information at input level. In this paper, we propose Dual Domain Unet (DuDoUnet) with specially designed Information Sharing Block (ISB), which can further share two domains' information and distinguish artifacts. Besides, we use mutual information (MI) with an auxiliary network, whose inputs and outputs are both ground truth, to compensate prior knowledge of limited-view PA inputs. The proposed method is verified with a public clinical database, and shows superior results with SSIM = 93.5622% and PSNR = 20.8859.

eess.IV↗

arXiv2020

Generalized Inverse Planning: Learning Lifted non-Markovian Utility for Generalizable Task Representation

In searching for a generalizable representation of temporally extended tasks, we spot two necessary constituents: the utility needs to be non-Markovian to transfer temporal relations invariant to a probability shift, the utility also needs to be lifted to abstract out specific grounding objects. In this work, we study learning such utility from human demonstrations. While inverse reinforcement learning (IRL) has been accepted as a general framework of utility learning, its fundamental formulation is one concrete Markov Decision Process. Thus the learned reward function does not specify the task independently of the environment. Going beyond that, we define a domain of generalization that spans a set of planning problems following a schema. We hence propose a new quest, Generalized Inverse Planning, for utility learning in this domain. We further outline a computational framework, Maximum Entropy Inverse Planning (MEIP), that learns non-Markovian utility and associated concepts in a generative manner. The learned utility and concepts form a task representation that generalizes regardless of probability shift or structural change. Seeing that the proposed generalization problem has not been widely studied yet, we carefully define an evaluation protocol, with which we illustrate the effectiveness of MEIP on two proof-of-concept domains and one challenging task: learning to fold from demonstrations.

cs.LG↗

arXiv2020

The role of charge recombination to spin-triplet excitons in non-fullerene acceptor organic solar cells

The power conversion efficiencies (PCEs) of organic solar cells (OSCs) using non-fullerene acceptors (NFAs) have now reached 18%. However, this is still lower than inorganic solar cells, for which PCEs >20% are commonplace. A key reason is that OSCs still show low open-circuit voltages (Voc) relative to their optical band gaps, attributed to non-radiative recombination. For OSCs to compete with inorganics in efficiency, all non-radiative loss pathways must be identified and where possible, removed. Here, we show that in most NFA OSCs, the majority of charge recombination at open-circuit proceeds via formation of non-emissive NFA triplet excitons (T1); in the benchmark PM6:Y6 blend, this fraction reaches 90%, contributing 60 mV to the reduction of Voc. We develop a new design to prevent recombination via this non-radiative channel through the engineering of significant hybridisation between the NFA T1 and the spin-triplet charge transfer exciton (3CTE). We model that the rate of the back charge transfer from 3CTE to T1 can be reduced by an order of magnitude, allowing re-dissociation of the 3CTE. We then demonstrate NFA systems where T1 formation is suppressed. This work therefore provides a clear design pathway for improved OSC performance to 20% PCE and beyond.

physics.app-ph↗

arXiv2020

Shelf-stable quantum-dot light-emitting diodes with high operational performance

Quantum-dot light-emitting diodes (QLEDs) promise high-performance and cost-effective electroluminescent devices. However, shelf stability of QLEDs, a must for practical applications, is currently overlooked. Here we reveal that the state-of-the-art QLEDs exhibit abnormal shelf-ageing behaviours, i.e., improvements in performance (positive ageing), followed by deterioration of performance (negative ageing). Mechanism studies show that the organic acids in the encapsulation acrylic resin induce in-situ reactions, which simultaneously improve electrical conductance and minimize interfacial exciton quenching at the positive-ageing stage. Progression of the in-situ reactions results in negative ageing. Inspired by these findings, we design an electron-transporting bi-layer structure, which delivers both improved electrical conductivity and suppressed interfacial exciton quenching. This design enables shelf-stable QLEDs with high operational performance, i.e., neglectable changes of external quantum efficiency (>20.0%) and ultra-long operational lifetime (T95: 5,500 hours at 1,000 cd m-2) after storage for 180 days. Our work paves the way towards the commercialization of QLEDs.

physics.app-ph↗

arXiv2020

Light-scanning hand-held photoacoustic probe design

Significance: We proposed a new design of hand-held linear-array photoacoustic (PA) probe which can acquire multi images via motor moving. Moreover, images from different locations are utilized via imaging fusion for SNR enhancement. Aim: We devised an adjustable hand-held for the purpose of realizing different images at diverse location for further image fusion. For realizing the light spot which is more matched with the Ultrasonic transducer detection area, we specially design a light adjust unit. Moreover, due to no displacement among the images, there is no need to execute image register process. The program execution time be reduced, greatly. Approach: mechanical design; Montel carol simulation; no-registration image fusion; Spot compression. Results: Multiple PA images with different optical illumination areas were acquired. After image fusion, we obtained fused PA images with higher signal-to-noise-ratio (SNR) and image fidelity than each single PA image. A quantitative comparison shows that the SNR of fused image is improved by 36.06% in agar-milk phantom, and 44.69% in chicken breast phantom, respectively. Conclusions: In this paper, the light scanning adjustable hand-held PA imaging probe is proposed, which can realize PA imaging with different illumination positions via adjusting the optical unit.

physics.med-ph↗

arXiv2020

Conceptualized Representation Learning for Chinese Biomedical Text Mining

Biomedical text mining is becoming increasingly important as the number of biomedical documents and web data rapidly grows. Recently, word representation models such as BERT has gained popularity among researchers. However, it is difficult to estimate their performance on datasets containing biomedical texts as the word distributions of general and biomedical corpora are quite different. Moreover, the medical domain has long-tail concepts and terminologies that are difficult to be learned via language models. For the Chinese biomedical text, it is more difficult due to its complex structure and the variety of phrase combinations. In this paper, we investigate how the recently introduced pre-trained language model BERT can be adapted for Chinese biomedical corpora and propose a novel conceptualized representation learning approach. We also release a new Chinese Biomedical Language Understanding Evaluation benchmark (\textbf{ChineseBLUE}). We examine the effectiveness of Chinese pre-trained models: BERT, BERT-wwm, RoBERTa, and our approach. Experimental results on the benchmark show that our approach could bring significant gain. We release the pre-trained model on GitHub: https://github.com/alibaba-research/ChineseBLUE.

cs.CL↗

arXiv2020

When Do Drivers Concentrate? Attention-based Driver Behavior Modeling With Deep Reinforcement Learning

Driver distraction a significant risk to driving safety. Apart from spatial domain, research on temporal inattention is also necessary. This paper aims to figure out the pattern of drivers' temporal attention allocation. In this paper, we propose an actor-critic method - Attention-based Twin Delayed Deep Deterministic policy gradient (ATD3) algorithm to approximate a driver' s action according to observations and measure the driver' s attention allocation for consecutive time steps in car-following model. Considering reaction time, we construct the attention mechanism in the actor network to capture temporal dependencies of consecutive observations. In the critic network, we employ Twin Delayed Deep Deterministic policy gradient algorithm (TD3) to address overestimated value estimates persisting in the actor-critic algorithm. We conduct experiments on real-world vehicle trajectory datasets and show that the accuracy of our proposed approach outperforms seven baseline algorithms. Moreover, the results reveal that the attention of the drivers in smooth vehicles is uniformly distributed in previous observations while they keep their attention to recent observations when sudden decreases of relative speeds occur. This study is the first contribution to drivers' temporal attention and provides scientific support for safety measures in transportation systems from the perspective of data mining.

cs.LG↗

arXiv2020

MEMS Heat Flux Sensor

Heat flux sensors have potential in enabling applications that require tracking of direct and instantaneous thermal energy transfer. To facilitate their use, the sensors have to be robust and feasible to implement, while maintaining high sensitivity and a fast response time. However, most commercially available heat flux sensors are expensive to manufacture and have insufficient temporal responses. In this paper, a novel microelectromechanical heat flux sensor structure is proposed. The electrical performance of the prototype sensors is compared with commercially available heat flux sensors. Preliminary results show that our sensors have similar sensitivity and faster response compared to commercial sensors.

physics.app-ph↗

arXiv2020

Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense

Recent progress in deep learning is essentially based on a "big data for small tasks" paradigm, under which massive amounts of data are used to train a classifier for a single narrow task. In this paper, we call for a shift that flips this paradigm upside down. Specifically, we propose a "small data for big tasks" paradigm, wherein a single artificial intelligence (AI) system is challenged to develop "common sense", enabling it to solve a wide range of tasks with little training data. We illustrate the potential power of this new paradigm by reviewing models of common sense that synthesize recent breakthroughs in both machine and human vision. We identify functionality, physics, intent, causality, and utility (FPICU) as the five core domains of cognitive AI with humanlike common sense. When taken as a unified concept, FPICU is concerned with the questions of "why" and "how", beyond the dominant "what" and "where" framework for understanding vision. They are invisible in terms of pixels but nevertheless drive the creation, maintenance, and development of visual scenes. We therefore coin them the "dark matter" of vision. Just as our universe cannot be understood by merely studying observable matter, we argue that vision cannot be understood without studying FPICU. We demonstrate the power of this perspective to develop cognitive AI systems with humanlike common sense by showing how to observe and apply FPICU with little training data to solve a wide range of challenging tasks, including tool use, planning, utility inference, and social learning. In summary, we argue that the next generation of AI must embrace "dark" humanlike common sense for solving novel tasks.

cs.AI↗

arXiv2020

A Generative Adversarial Network for AI-Aided Chair Design

We present a method for improving human design of chairs. The goal of the method is generating enormous chair candidates in order to facilitate human designer by creating sketches and 3d models accordingly based on the generated chair design. It consists of an image synthesis module, which learns the underlying distribution of training dataset, a super-resolution module, which improve quality of generated image and human involvements. Finally, we manually pick one of the generated candidates to create a real life chair for illustration.

cs.CV↗

arXiv2020

Multilayer InSe-Te van der Waals heterostructures with ultrahigh rectification ratio and ultrasensitive photoresponse

Multilayer van der Waals (vdWs) semiconductors have great promising application in high-performance optoelectronic devices. However, the photoconductive photodetectors based on layered semiconductors often suffer from large dark current and high external driven bias voltage. Here, we report a vertical van der Waals heterostructures (vdWHs) consisting of multilayer indium selenide (InSe) and tellurium (Te). The multilayer InSe-Te vdWHs device shows a record high forward rectification ratio greater than 107 at room temperature. Furthermore, an ultrasensitive and broadband photoresponse photodetector is achieved by the vdWHs device with an ultrahigh photo/dark current ratio over 104, a high detectivity of 1013, and a comparable responsivity of 0.45 A/W under visible light illumination with weak incident power. Moreover, the vdWHs device has a photovoltaic effect and can function as a self-powered photodetector (SPPD). The SPPD is also ultrasensitive to the broadband spectra ranging from 300 nm to 1000 nm and is capable of detecting weak light signals. This work offers an opportunity to develop next-generation electronic and optoelectronic devices based on multilayer vdWs structures.

physics.app-ph↗

arXiv2019

Frequency tunable topological edge states of two-dimensional honeycomb lattice photonic crystals

In this paper, the photonic quantum spin Hall effect (PQSHE) is realized in dielectric two-dimensional (2D) honeycomb lattice photonic crystal (PC) by stretching and shrinking the honeycomb unit cell. Combining two honeycomb lattice PCs with a common photonic band gap (PBG) but different band topologies can generate a topologically protected edge state at the combined junction. The topological edge states and their unidirectional transmission as the scatterers with triangular, pentagonal, and heptagonal shapes are researched. Meanwhile, the unidirectional transmission in an inverted Ω-shaped waveguide with large bending angle is realized, and verifies the characteristics of the topological protection by adding different kind of defects. Moreover, the frequency varies significantly when changing the scatterers shape, which shows that the PC with various scatterers shape can tune the frequency range of the topological edge state significantly. In other words, it can adjust the frequency of unidirectional transmission and increase the adjustability of the topological edge state.

physics.optics↗

arXiv2019

Learning Perceptual Inference by Contrasting

"Thinking in pictures," [1] i.e., spatial-temporal reasoning, effortless and instantaneous for humans, is believed to be a significant ability to perform logical induction and a crucial factor in the intellectual history of technology development. Modern Artificial Intelligence (AI), fueled by massive datasets, deeper models, and mighty computation, has come to a stage where (super-)human-level performances are observed in certain specific tasks. However, current AI's ability in "thinking in pictures" is still far lacking behind. In this work, we study how to improve machines' reasoning ability on one challenging task of this kind: Raven's Progressive Matrices (RPM). Specifically, we borrow the very idea of "contrast effects" from the field of psychology, cognition, and education to design and train a permutation-invariant model. Inspired by cognitive studies, we equip our model with a simple inference module that is jointly trained with the perception backbone. Combining all the elements, we propose the Contrastive Perceptual Inference network (CoPINet) and empirically demonstrate that CoPINet sets the new state-of-the-art for permutation-invariant models on two major datasets. We conclude that spatial-temporal reasoning depends on envisaging the possibilities consistent with the relations between objects and can be solved from pixel-level inputs.

cs.CV↗

arXiv2019

Spatiotemporal Attention Networks for Wind Power Forecasting

Wind power is one of the most important renewable energy sources and accurate wind power forecasting is very significant for reliable and economic power system operation and control strategies. This paper proposes a novel framework with spatiotemporal attention networks (STAN) for wind power forecasting. This model captures spatial correlations among wind farms and temporal dependencies of wind power time series. First of all, we employ a multi-head self-attention mechanism to extract spatial correlations among wind farms. Then, temporal dependencies are captured by the Sequence-to-Sequence (Seq2Seq) model with a global attention mechanism. Finally, experimental results demonstrate that our model achieves better performance than other baseline approaches. Our work provides useful insights to capture non-Euclidean spatial correlations.

cs.LG↗

arXiv2019

A Simulation Approach to Multi-station Solar Irradiance Data Considering Temporal Correlations

Solar energy is one of important renewable energy sources and simulation of solar irradiance can be used as input for simulation of photovoltaic (PV) generation. This paper proposes a simulation algorithm of multi-station solar irradiance data considering temporal correlations. First of all, we group all the days of the observed data to k clusters for each station based on their daily features of solar irradiance and the daily states constitute Markov chain of days. Then, we reduce state permutations of different stations before getting Markov Transition Probability Matrix (MTPM). In terms of the observed data and MTPM, the simulation approach is proposed. Finally, we test our approach by applying to solar irradiance data of three stations and show that the properties of simulated data match those of the observed data.

eess.SP↗

arXiv2019

Perfect Optical Nonreciprocity in a Double-Cavity Optomechanical System

Nonreciprocal devices are indispensable for building quantum networks and ubiquitous in modern communication technology. Here, we use optomechanical interaction and linearly-coupled interaction to realize optical nonreciprocal transmission in a double-cavity optomechanical system. The scheme relies on the interference between the two interactions. We derive the essential conditions to realize perfect optical nonreciprocity in the system, and analyse the properties of optical nonreciprocal transmission and the output fields from mechanical mode. These results can be used to control optical transmission in quantum information processing.

quant-ph↗

arXiv2019

Scalar field effects on the orbit of S2 star

Precise measurements of the S-stars orbiting SgrA* have set strong constraints on the nature of the compact object at the centre of the Milky Way. The presence of a black hole in that region is well established, but its neighboring environment is still an open debate. In that respect, the existence of dark matter in that central region may be detectable due to its strong signatures on the orbits of stars: the main effect is a Newtonian precession which will affect the overall pericentre shift of S2, the latter being a target measurement of the GRAVITY instrument. The exact nature of this dark matter (e.g., stellar dark remnants or diffuse dark matter) is unknown. This article assumes it to be an scalar field of toroidal distribution, associated with ultra-light dark matter particles, surrounding the Kerr black hole. Such a field is a form of "hair" expected in the context of superradiance, a mechanism that extracts rotational energy from the black hole. Orbital signatures for the S2 star are computed and shown to be detectable by GRAVITY. The scalar field can be constrained because the variation of orbital elements depends both on the relative mass of the scalar field to the black hole and on the field mass coupling parameter.

astro-ph.GA↗

arXiv2019

VRGym: A Virtual Testbed for Physical and Interactive AI

We propose VRGym, a virtual reality testbed for realistic human-robot interaction. Different from existing toolkits and virtual reality environments, the VRGym emphasizes on building and training both physical and interactive agents for robotics, machine learning, and cognitive science. VRGym leverages mechanisms that can generate diverse 3D scenes with high realism through physics-based simulation. We demonstrate that VRGym is able to (i) collect human interactions and fine manipulations, (ii) accommodate various robots with a ROS bridge, (iii) support experiments for human-robot interaction, and (iv) provide toolkits for training the state-of-the-art machine learning algorithms. We hope VRGym can help to advance general-purpose robotics and machine learning agents, as well as assisting human studies in the field of cognitive science.

cs.HC↗