arXiv subjects

Yang Hu

Publications and source records attributed to Yang Hu.

At least 181 records · Page 10Linked to original sources

arXiv2021

Recent Advances on Non-Line-of-Sight Imaging: Conventional Physical Models, Deep Learning, and New Scenes

As an emerging technology that has attracted huge attention, non-line-of-sight (NLOS) imaging can reconstruct hidden objects by analyzing the diffuse reflection on a relay surface, with broad application prospects in the fields of autonomous driving, medical imaging, and defense. Despite the challenges of low signal-to-noise ratio (SNR) and high ill-posedness, NLOS imaging has been developed rapidly in recent years. Most current NLOS imaging technologies use conventional physical models, constructing imaging models through active or passive illumination and using reconstruction algorithms to restore hidden scenes. Moreover, deep learning algorithms for NLOS imaging have also received much attention recently. This paper presents a comprehensive overview of both conventional and deep learning-based NLOS imaging techniques. Besides, we also survey new proposed NLOS scenes, and discuss the challenges and prospects of existing technologies. Such a survey can help readers have an overview of different types of NLOS imaging, thus expediting the development of seeing around corners.

eess.IV↗

arXiv2021

Calculus of Consent via MARL: Legitimating the Collaborative Governance Supplying Public Goods

Public policies that supply public goods, especially those involve collaboration by limiting individual liberty, always give rise to controversies over governance legitimacy. Multi-Agent Reinforcement Learning (MARL) methods are appropriate for supporting the legitimacy of the public policies that supply public goods at the cost of individual interests. Among these policies, the inter-regional collaborative pandemic control is a prominent example, which has become much more important for an increasingly inter-connected world facing a global pandemic like COVID-19. Different patterns of collaborative strategies have been observed among different systems of regions, yet it lacks an analytical process to reason for the legitimacy of those strategies. In this paper, we use the inter-regional collaboration for pandemic control as an example to demonstrate the necessity of MARL in reasoning, and thereby legitimizing policies enforcing such inter-regional collaboration. Experimental results in an exemplary environment show that our MARL approach is able to demonstrate the effectiveness and necessity of restrictions on individual liberty for collaborative supply of public goods. Different optimal policies are learned by our MARL agents under different collaboration levels, which change in an interpretable pattern of collaboration that helps to balance the losses suffered by regions of different types, and consequently promotes the overall welfare. Meanwhile, policies learned with higher collaboration levels yield higher global rewards, which illustrates the benefit of, and thus provides a novel justification for the legitimacy of, promoting inter-regional collaboration. Therefore, our method shows the capability of MARL in computationally modeling and supporting the theory of calculus of consent, developed by Nobel Prize winner J. M. Buchanan.

cs.AI↗

arXiv2021

Rotation-induced significant modulation of near-field radiative heat transfer between hyperbolic nanoparticles

Modulation of near-field radiative heat transfer (NFRHT) with rotated anisotropic hBN and MoO3 nanoparticles (NPs) are investigated. The spectral heat power, total heat power and electric field energy density are calculated at different particle orientations. Numerical results show that the modulation factor of the MoO3 NPs could be up to ~12000 with particle radius of 40 nm at a gap distance of 200 nm, which is ~ 10-fold larger than the state of the art. Excitation of localized hyperbolic phonon polaritons (LHPPs) in different particle orientations allow for the excellent modulation. The modulation factor of the hBN NPs is 5.9 due to the insensitivity of the LHPPs in the type II hyperbolic band to the particle orientations. This work not only provides a new method to modulate NFRHT in particle systems, but also paves the way to explore radiative heat transfer characteristics of anisotropic media.

physics.optics↗

arXiv2021

Pipeline Parallelism for Inference on Heterogeneous Edge Computing

Deep neural networks with large model sizes achieve state-of-the-art results for tasks in computer vision (CV) and natural language processing (NLP). However, these large-scale models are too compute- or memory-intensive for resource-constrained edge devices. Prior works on parallel and distributed execution primarily focus on training -- rather than inference -- using homogeneous accelerators in data centers. We propose EdgePipe, a distributed framework for edge systems that uses pipeline parallelism to both speed up inference and enable running larger (and more accurate) models that otherwise cannot fit on single edge devices. EdgePipe achieves these results by using an optimal partition strategy that considers heterogeneity in compute, memory, and network bandwidth. Our empirical evaluation demonstrates that EdgePipe achieves $10.59\times$ and $11.88\times$ speedup using 16 edge devices for the ViT-Large and ViT-Huge models, respectively, with no accuracy loss. Similarly, EdgePipe improves ViT-Huge throughput by $3.93\times$ over a 4-node baseline using 16 edge devices, which independently cannot fit the model in memory. Finally, we show up to $4.16\times$ throughput improvement over the state-of-the-art PipeDream when using a heterogeneous set of devices.

cs.DC↗

arXiv2021

Deep Reinforcement Learning with Modulated Hebbian plus Q Network Architecture

This paper presents a new neural architecture that combines a modulated Hebbian network (MOHN) with DQN, which we call modulated Hebbian plus Q network architecture (MOHQA). The hypothesis is that such a combination allows MOHQA to solve difficult partially observable Markov decision process (POMDP) problems which impair temporal difference (TD)-based RL algorithms such as DQN, as the TD error cannot be easily derived from observations. The key idea is to use a Hebbian network with bio-inspired neural traces in order to bridge temporal delays between actions and rewards when confounding observations and sparse rewards result in inaccurate TD errors. In MOHQA, DQN learns low level features and control, while the MOHN contributes to the high-level decisions by associating rewards with past states and actions. Thus the proposed architecture combines two modules with significantly different learning algorithms, a Hebbian associative network and a classical DQN pipeline, exploiting the advantages of both. Simulations on a set of POMDPs and on the MALMO environment show that the proposed algorithm improved DQN's results and even outperformed control tests with A2C, QRDQN+LSTM and REINFORCE algorithms on some POMDPs with confounding stimuli and sparse rewards.

cs.LG↗

arXiv2021

Data-Driven Electron Microscopy: Electron Diffraction Imaging of Materials Structural Properties

Transmission electron diffraction is a powerful and versatile structural probe for the characterization of a broad range of materials, from nanocrystalline thin films to single crystals. With recent developments in fast electron detectors and efficient computer algorithms, it now becomes possible to collect unprecedently large datasets of diffraction patterns (DPs) and process DPs to extract crystallographic information to form images or tomograms based on crystal structural properties, giving rise to data-driven electron microscopy. Critical to this kind of imaging is the type of crystallographic information being collected, which can be achieved with a judicious choice of electron diffraction techniques, and the efficiency and accuracy of DP processing, which requires the development of new algorithms. Here, we review recent progress made in data collection, new algorithms, and automated electron DP analysis. These progresses will be highlighted using application examples in materials research. Future opportunities based on smart sampling and machine learning are also discussed.

cond-mat.mtrl-sci↗

arXiv2021

Inner-Imaging Networks: Put Lenses into Convolutional Structure

Despite the tremendous success in computer vision, deep convolutional networks suffer from serious computation costs and redundancies. Although previous works address this issue by enhancing diversities of filters, they have not considered the complementarity and the completeness of the internal structure of the convolutional network. To deal with these problems, a novel Inner-Imaging architecture is proposed in this paper, which allows relationships between channels to meet the above requirement. Specifically, we organize the channel signal points in groups using convolutional kernels to model both the intra-group and inter-group relationships simultaneously. The convolutional filter is a powerful tool for modeling spatial relations and organizing grouped signals, so the proposed methods map the channel signals onto a pseudo-image, like putting a lens into convolution internal structure. Consequently, not only the diversity of channels is increased, but also the complementarity and completeness can be explicitly enhanced. The proposed architecture is lightweight and easy to be implemented. It provides an efficient self-organization strategy for convolutional networks so as to improve their efficiency and performance. Extensive experiments are conducted on multiple benchmark image recognition data sets including CIFAR, SVHN and ImageNet. Experimental results verify the effectiveness of the Inner-Imaging mechanism with the most popular convolutional networks as the backbones.

cs.CV↗

arXiv2021

Heredity-aware Child Face Image Generation with Latent Space Disentanglement

Generative adversarial networks have been widely used in image synthesis in recent years and the quality of the generated image has been greatly improved. However, the flexibility to control and decouple facial attributes (e.g., eyes, nose, mouth) is still limited. In this paper, we propose a novel approach, called ChildGAN, to generate a child's image according to the images of parents with heredity prior. The main idea is to disentangle the latent space of a pre-trained generation model and precisely control the face attributes of child images with clear semantics. We use distances between face landmarks as pseudo labels to figure out the most influential semantic vectors of the corresponding face attributes by calculating the gradient of latent vectors to pseudo labels. Furthermore, we disentangle the semantic vectors by weighting irrelevant features and orthogonalizing them with Schmidt Orthogonalization. Finally, we fuse the latent vector of the parents by leveraging the disentangled semantic vectors under the guidance of biological genetic laws. Extensive experiments demonstrate that our approach outperforms the existing methods with encouraging results.

cs.CV↗

arXiv2021

Radiative thermal switch exploiting hyperbolic surface phonon polaritons

We study the radiative heat flux between two nanoparticles in close vicinity to the natural hyperbolic material hBN with its optical axis oriented parallel to the interface. We show that the heat flux between the nanoparticles can be efficiently modulated when rotating the nanoparticles due to the coupling to the highly directional hyperbolic surface modes in hBN. Finally, we discuss the thickness and distance dependence of this effect.

cond-mat.mes-hall↗

arXiv2021

Perturbation-based Regret Analysis of Predictive Control in Linear Time Varying Systems

We study predictive control in a setting where the dynamics are time-varying and linear, and the costs are time-varying and well-conditioned. At each time step, the controller receives the exact predictions of costs, dynamics, and disturbances for the future $k$ time steps. We show that when the prediction window $k$ is sufficiently large, predictive control is input-to-state stable and achieves a dynamic regret of $O(λ^k T)$, where $λ< 1$ is a positive constant. This is the first dynamic regret bound on the predictive control of linear time-varying systems. Under more assumptions on the terminal costs, we also show that predictive control obtains the first competitive bound for the control of linear time-varying systems: $1 + O(λ^k)$. Our results are derived using a novel proof framework based on a perturbation bound that characterizes how a small change to the system parameters impacts the optimal trajectory.

math.OC↗

arXiv2021

Graph-based Visual-Semantic Entanglement Network for Zero-shot Image Recognition

Zero-shot learning uses semantic attributes to connect the search space of unseen objects. In recent years, although the deep convolutional network brings powerful visual modeling capabilities to the ZSL task, its visual features have severe pattern inertia and lack of representation of semantic relationships, which leads to severe bias and ambiguity. In response to this, we propose the Graph-based Visual-Semantic Entanglement Network to conduct graph modeling of visual features, which is mapped to semantic attributes by using a knowledge graph, it contains several novel designs: 1. it establishes a multi-path entangled network with the convolutional neural network (CNN) and the graph convolutional network (GCN), which input the visual features from CNN to GCN to model the implicit semantic relations, then GCN feedback the graph modeled information to CNN features; 2. it uses attribute word vectors as the target for the graph semantic modeling of GCN, which forms a self-consistent regression for graph modeling and supervise GCN to learn more personalized attribute relations; 3. it fuses and supplements the hierarchical visual-semantic features refined by graph modeling into visual embedding. Our method outperforms state-of-the-art approaches on multiple representative ZSL datasets: AwA2, CUB, and SUN by promoting the semantic linkage modelling of visual features.

cs.CV↗

arXiv2021

What Can Knowledge Bring to Machine Learning? -- A Survey of Low-shot Learning for Structured Data

Supervised machine learning has several drawbacks that make it difficult to use in many situations. Drawbacks include: heavy reliance on massive training data, limited generalizability and poor expressiveness of high-level semantics. Low-shot Learning attempts to address these drawbacks. Low-shot learning allows the model to obtain good predictive power with very little or no training data, where structured knowledge plays a key role as a high-level semantic representation of human. This article will review the fundamental factors of low-shot learning technologies, with a focus on the operation of structured knowledge under different low-shot conditions. We also introduce other techniques relevant to low-shot learning. Finally, we point out the limitations of low-shot learning, the prospects and gaps of industrial applications, and future research directions.

cs.LG↗

arXiv2021

Graph-MLP: Node Classification without Message Passing in Graph

Graph Neural Network (GNN) has been demonstrated its effectiveness in dealing with non-Euclidean structural data. Both spatial-based and spectral-based GNNs are relying on adjacency matrix to guide message passing among neighbors during feature aggregation. Recent works have mainly focused on powerful message passing modules, however, in this paper, we show that none of the message passing modules is necessary. Instead, we propose a pure multilayer-perceptron-based framework, Graph-MLP with the supervision signal leveraging graph structure, which is sufficient for learning discriminative node representation. In model-level, Graph-MLP only includes multi-layer perceptrons, activation function, and layer normalization. In the loss level, we design a neighboring contrastive (NContrast) loss to bridge the gap between GNNs and MLPs by utilizing the adjacency information implicitly. This design allows our model to be lighter and more robust when facing large-scale graph data and corrupted adjacency information. Extensive experiments prove that even without adjacency information in testing phase, our framework can still reach comparable and even superior performance against the state-of-the-art models in the graph node classification task.

cs.LG↗

arXiv2021

Tackling Variabilities in Autonomous Driving

The state-of-the-art driving automation system demands extreme computational resources to meet rigorous accuracy and latency requirements. Though emerging driving automation computing platforms are based on ASIC to provide better performance and power guarantee, building such an accelerator-based computing platform for driving automation still present challenges. First, the workloads mix and performance requirements exposed to driving automation system present significant variability. Second, with more cameras/sensors integrated in a future fully autonomous driving vehicle, a heterogeneous multi-accelerator architecture substrate is needed that requires a design space exploration for a new form of parallelism. In this work, we aim to extensively explore the above system design challenges and these challenges motivate us to propose a comprehensive framework that synergistically handles the heterogeneous hardware accelerator design principles, system design criteria, and task scheduling mechanism. Specifically, we propose a novel heterogeneous multi-core AI accelerator (HMAI) to provide the hardware substrate for the driving automation tasks with variability. We also define system design criteria to better utilize hardware resources and achieve increased throughput while satisfying the performance and energy restrictions. Finally, we propose a deep reinforcement learning (RL)-based task scheduling mechanism FlexAI, to resolve task mapping issue. Experimental results show that with FlexAI scheduling, basically 100% tasks in each driving route can be processed by HMAI within their required period to ensure safety, and FlexAI can also maximally reduce the breaking distance up to 96% as compared to typical heuristics and guided random-search-based algorithms.

cs.AR↗

arXiv2021

Q-VR: System-Level Design for Future Mobile Collaborative Virtual Reality

High Quality Mobile Virtual Reality (VR) is what the incoming graphics technology era demands: users around the world, regardless of their hardware and network conditions, can all enjoy the immersive virtual experience. However, the state-of-the-art software-based mobile VR designs cannot fully satisfy the realtime performance requirements due to the highly interactive nature of user's actions and complex environmental constraints during VR execution. Inspired by the unique human visual system effects and the strong correlation between VR motion features and realtime hardware-level information, we propose Q-VR, a novel dynamic collaborative rendering solution via software-hardware co-design for enabling future low-latency high-quality mobile VR. At software-level, Q-VR provides flexible high-level tuning interface to reduce network latency while maintaining user perception. At hardware-level, Q-VR accommodates a wide spectrum of hardware and network conditions across users by effectively leveraging the computing capability of the increasingly powerful VR hardware. Extensive evaluation on real-world games demonstrates that Q-VR can achieve an average end-to-end performance speedup of 3.4x (up to 6.7x) over the traditional local rendering design in commercial VR devices, and a 4.1x frame rate improvement over the state-of-the-art static collaborative rendering.

cs.AR↗

arXiv2021

Cepstral Scanning Transmission Electron Microscopy Imaging of Severe Lattice Distortions

The development of four-dimensional (4D) scanning transmission electron microscopy (STEM) using fast detectors has opened-up new avenues for addressing some of long-standing challenges in electron imaging. One of these challenges is how to image severely distorted crystal lattices, such as at a dislocation core. Here we introduce a new 4D-STEM technique, called Cepstral STEM, for imaging disordered crystals using electron diffuse scattering. Local fluctuations of diffuse scattering are captured by scanning electron nanodiffraction (SEND) using a coherent probe. The harmonic signals in electron diffuse scattering are detected through Cepstral analysis and used for imaging. By integrating Cepstral analysis with 4D-STEM, we demonstrate that information about the distortive part of electron scattering potential can be separated and imaged at nm spatial resolution. We apply our technique to the analysis of a dislocation core in SiGe and lattice distortions in high entropy alloy.

cond-mat.mtrl-sci↗

arXiv2020

Embracing the chaos: Alloying adds stochasticity to twin embryo growth

High-throughput atomistic simulations reveal the unique effect of solute atoms on twin variant selection in Mg-Al alloys. Twin embryos first undergo a stochastic incubation stage when embryos choose which twin variant to adopt, and then a deterministic growth stage when embryos expand without changing the selected twin variant. An increase in Al composition promotes the stochastic incubation behavior on the atomic level due to nucleation and pinning of interfacial disconnections. At compositions above a critical value, disconnection pinning results in multiple twin variant selection.

cond-mat.mtrl-sci↗

arXiv2020

Nanoscale control of LaAlO3/SrTiO3 metal-insulator transition using ultra-low-voltage electron-beam lithography

We describe a method to control the insulator-metal transition at the LaAlO3/SrTiO3 interface using ultra-low-voltage electron beam lithography (ULV-EBL). Compared with previous reports that utilize conductive atomic-force-microscope lithography (c-AFM), this approach can provide comparable resolution (~10 nm) at write speeds (10 mm/s) that are up to 10,000x faster than c-AFM. The writing technique is non-destructive and the conductive state is reversible via prolonged exposure to air. Transport properties of representative devices are measured at milli-Kelvin temperatures, where superconducting behavior is observed. We also demonstrate the ability to create conducting devices on graphene/LaAlO3/SrTiO3 heterostructures. The underlying mechanism is believed to be closely related to the same mechanism regulating c-AFM-based methods.

cond-mat.mes-hall↗