arXiv subjects

Chong Huang

Publications and source records attributed to Chong Huang.

35 records · Page 2Linked to original sources

arXiv2024

Joint Offloading and Resource Allocation for Hybrid Cloud and Edge Computing in SAGINs: A Decision Assisted Hybrid Action Space Deep Reinforcement Learning Approach

In recent years, the amalgamation of satellite communications and aerial platforms into space-air-ground integrated network (SAGINs) has emerged as an indispensable area of research for future communications due to the global coverage capacity of low Earth orbit (LEO) satellites and the flexible Deployment of aerial platforms. This paper presents a deep reinforcement learning (DRL)-based approach for the joint optimization of offloading and resource allocation in hybrid cloud and multi-access edge computing (MEC) scenarios within SAGINs. The proposed system considers the presence of multiple satellites, clouds and unmanned aerial vehicles (UAVs). The multiple tasks from ground users are modeled as directed acyclic graphs (DAGs). With the goal of reducing energy consumption and latency in MEC, we propose a novel multi-agent algorithm based on DRL that optimizes both the offloading strategy and the allocation of resources in the MEC infrastructure within SAGIN. A hybrid action algorithm is utilized to address the challenge of hybrid continuous and discrete action space in the proposed problems, and a decision-assisted DRL method is adopted to reduce the impact of unavailable actions in the training process of DRL. Through extensive simulations, the results demonstrate the efficacy of the proposed learning-based scheme, the proposed approach consistently outperforms benchmark schemes, highlighting its superior performance and potential for practical applications.

cs.IT↗

arXiv2024

Deep Learning Driven Buffer-Aided Cooperative Networks for B5G/6G: Challenges, Solutions, and Future Opportunities

Buffer-aided cooperative networks (BACNs) have garnered significant attention due to their potential applications in beyond fifth generation (B5G) or sixth generation (6G) critical scenarios. This article explores various typical application scenarios of buffer-aided relaying in B5G/6G networks to emphasize the importance of incorporating BACN. Additionally, we delve into the crucial technical challenges in BACN, including stringent delay constraints, high reliability, imperfect channel state information (CSI), transmission security, and integrated network architecture. To address the challenges, we propose leveraging deep learning-based methods for the design and operation of B5G/6G networks with BACN, deviating from conventional buffer-aided relay selection approaches. In particular, we present two case studies to demonstrate the efficacy of centralized deep reinforcement learning (DRL) and decentralized DRL in buffer-aided non-terrestrial networks. Finally, we outline future research directions in B5G/6G that pertain to the utilization of BACN.

cs.IT↗

arXiv2023

Time-resolved laser speckle contrast imaging (TR-LSCI) of cerebral blood flow

To address many of the deficiencies in optical neuroimaging technologies such as poor spatial resolution, time-consuming reconstruction, low penetration depth, and contact-based measurement, a novel, noncontact, time-resolved laser speckle contrast imaging (TR-LSCI) technique has been developed for continuous, fast, and high-resolution 2D mapping of cerebral blood flow (CBF) at different depths of the head. TR-LSCI illuminates the head with picosecond-pulsed, coherent, widefield near-infrared light and synchronizes a newly developed, high-resolution, gated single-photon avalanche diode camera (SwissSPAD2) to capture CBF maps at different depths. By selectively collecting diffuse photons with longer pathlengths through the head, TR-LSCI reduces partial volume artifacts from the overlying tissues, thus improving the accuracy of CBF measurement in the deep brain. CBF map reconstruction was dramatically expedited by incorporating highly parallelized computation. The performance of TR-LSCI was evaluated using head-simulating phantoms with known properties and in-vivo rodents with varied hemodynamic challenges to the brain. Results from these pilot studies demonstrated that TR-LSCI enabled mapping CBF variations at different depths with a sampling rate of up to 1 Hz and spatial resolutions ranging from tens of micrometers on the head surface to 1-2 millimeters in the deep brain. With additional improvements and validation in larger populations against established methods, we anticipate offering a noncontact, fast, high-resolution, portable, and affordable brain imager for fundamental neuroscience research in animals and for translational studies in humans.

physics.med-ph↗

arXiv2022

Learning Prototype via Placeholder for Zero-shot Recognition

Zero-shot learning (ZSL) aims to recognize unseen classes by exploiting semantic descriptions shared between seen classes and unseen classes. Current methods show that it is effective to learn visual-semantic alignment by projecting semantic embeddings into the visual space as class prototypes. However, such a projection function is only concerned with seen classes. When applied to unseen classes, the prototypes often perform suboptimally due to domain shift. In this paper, we propose to learn prototypes via placeholders, termed LPL, to eliminate the domain shift between seen and unseen classes. Specifically, we combine seen classes to hallucinate new classes which play as placeholders of the unseen classes in the visual and semantic space. Placed between seen classes, the placeholders encourage prototypes of seen classes to be highly dispersed. And more space is spared for the insertion of well-separated unseen ones. Empirically, well-separated prototypes help counteract visual-semantic misalignment caused by domain shift. Furthermore, we exploit a novel semantic-oriented fine-tuning to guarantee the semantic reliability of placeholders. Extensive experiments on five benchmark datasets demonstrate the significant performance gain of LPL over the state-of-the-art methods. Code is available at https://github.com/zaiquanyang/LPL.

cs.CV↗

arXiv2022

Generating Fair Universal Representations using Adversarial Models

We present a data-driven framework for learning fair universal representations (FUR) that guarantee statistical fairness for any learning task that may not be known a priori. Our framework leverages recent advances in adversarial learning to allow a data holder to learn representations in which a set of sensitive attributes are decoupled from the rest of the dataset. We formulate this as a constrained minimax game between an encoder and an adversary where the constraint ensures a measure of usefulness (utility) of the representation. The resulting problem is that of censoring, i.e., finding a representation that is least informative about the sensitive attributes given a utility constraint. For appropriately chosen adversarial loss functions, our censoring framework precisely clarifies the optimal adversarial strategy against strong information-theoretic adversaries; it also achieves the fairness measure of demographic parity for the resulting constrained representations. We evaluate the performance of our proposed framework on both synthetic and publicly available datasets. For these datasets, we use two tradeoff measures: censoring vs. representation fidelity and fairness vs. utility for downstream tasks, to amply demonstrate that multiple sensitive features can be effectively censored even as the resulting fair representations ensure accuracy for multiple downstream tasks.

cs.LG↗

arXiv2021

Hidden Technical Debts for Fair Machine Learning in Financial Services

The recent advancements in machine learning (ML) have demonstrated the potential for providing a powerful solution to build complex prediction systems in a short time. However, in highly regulated industries, such as the financial technology (Fintech), people have raised concerns about the risk of ML systems discriminating against specific protected groups or individuals. To address these concerns, researchers have introduced various mathematical fairness metrics and bias mitigation algorithms. This paper discusses hidden technical debts and challenges of building fair ML systems in a production environment for Fintech. We explore various stages that require attention for fairness in the ML system development and deployment life cycle. To identify hidden technical debts that exist in building fair ML system for Fintech, we focus on key pipeline stages including data preparation, model development, system monitoring and integration in production. Our analysis shows that enforcing fairness for production-ready ML systems in Fintech requires specific engineering commitments at different stages of ML system life cycle. We also propose several initial starting points to mitigate these technical debts for deploying fair ML systems in production.

cs.LG↗

arXiv2020

Delay Constrained Buffer-Aided Relay Selection in the Internet of Things with Decision-Assisted Reinforcement Learning

This paper investigates the reinforcement learning for the relay selection in the delay-constrained buffer-aided networks. The buffer-aided relay selection significantly improves the outage performance but often at the price of higher latency. On the other hand, modern communication systems such as the Internet of Things often have strict requirement on the latency. It is thus necessary to find relay selection policies to achieve good throughput performance in the buffer-aided relay network while stratifying the delay constraint. With the buffers employed at the relays and delay constraints imposed on the data transmission, obtaining the best relay selection becomes a complicated high-dimensional problem, making it hard for the reinforcement learning to converge. In this paper, we propose the novel decision-assisted deep reinforcement learning to improve the convergence. This is achieved by exploring the a-priori information from the buffer-aided relay system. The proposed approaches can achieve high throughput subject to delay constraints. Extensive simulation results are provided to verify the proposed algorithms.

eess.SP↗

arXiv2019

One-Shot Imitation Filming of Human Motion Videos

Imitation learning has been applied to mimic the operation of a human cameraman in several autonomous cinematography systems. To imitate different filming styles, existing methods train multiple models, where each model handles a particular style and requires a significant number of training samples. As a result, existing methods can hardly generalize to unseen styles. In this paper, we propose a framework, which can imitate a filming style by "seeing" only a single demonstration video of the same style, i.e., one-shot imitation filming. This is done by two key enabling techniques: 1) feature extraction of the filming style from the demo video, and 2) filming style transfer from the demo video to the new situation. We implement the approach with deep neural network and deploy it to a 6 degrees of freedom (DOF) real drone cinematography system by first predicting the future camera motions, and then converting them to the drone's control commands via an odometer. Our experimental results on extensive datasets and showcases exhibit significant improvements in our approach over conventional baselines and our approach can successfully mimic the footage with an unseen style.

cs.CV↗

arXiv2019

Generative Adversarial Privacy

We present a data-driven framework called generative adversarial privacy (GAP). Inspired by recent advancements in generative adversarial networks (GANs), GAP allows the data holder to learn the privatization mechanism directly from the data. Under GAP, finding the optimal privacy mechanism is formulated as a constrained minimax game between a privatizer and an adversary. We show that for appropriately chosen adversarial loss functions, GAP provides privacy guarantees against strong information-theoretic adversaries. We also evaluate GAP's performance on the GENKI face database.

cs.LG↗

arXiv2017

REDBEE: A Visual-Inertial Drone System for Real-Time Moving Object Detection

Aerial surveillance and monitoring demand both real-time and robust motion detection from a moving camera. Most existing techniques for drones involve sending a video data streams back to a ground station with a high-end desktop computer or server. These methods share one major drawback: data transmission is subjected to considerable delay and possible corruption. Onboard computation can not only overcome the data corruption problem but also increase the range of motion. Unfortunately, due to limited weight-bearing capacity, equipping drones with computing hardware of high processing capability is not feasible. Therefore, developing a motion detection system with real-time performance and high accuracy for drones with limited computing power is highly desirable. In this paper, we propose a visual-inertial drone system for real-time motion detection, namely REDBEE, that helps overcome challenges in shooting scenes with strong parallax and dynamic background. REDBEE, which can run on the state-of-the-art commercial low-power application processor (e.g. Snapdragon Flight board used for our prototype drone), achieves real-time performance with high detection accuracy. The REDBEE system overcomes obstacles in shooting scenes with strong parallax through an inertial-aided dual-plane homography estimation; it solves the issues in shooting scenes with dynamic background by distinguishing the moving targets through a probabilistic model based on spatial, temporal, and entropy consistency. The experiments are presented which demonstrate that our system obtains greater accuracy when detecting moving targets in outdoor environments than the state-of-the-art real-time onboard detection systems.

cs.RO↗

arXiv2017

Context-Aware Generative Adversarial Privacy

Preserving the utility of published datasets while simultaneously providing provable privacy guarantees is a well-known challenge. On the one hand, context-free privacy solutions, such as differential privacy, provide strong privacy guarantees, but often lead to a significant reduction in utility. On the other hand, context-aware privacy solutions, such as information theoretic privacy, achieve an improved privacy-utility tradeoff, but assume that the data holder has access to dataset statistics. We circumvent these limitations by introducing a novel context-aware privacy framework called generative adversarial privacy (GAP). GAP leverages recent advancements in generative adversarial networks (GANs) to allow the data holder to learn privatization schemes from the dataset itself. Under GAP, learning the privacy mechanism is formulated as a constrained minimax game between two players: a privatizer that sanitizes the dataset in a way that limits the risk of inference attacks on the individuals' private variables, and an adversary that tries to infer the private variables from the sanitized dataset. To evaluate GAP's performance, we investigate two simple (yet canonical) statistical dataset models: (a) the binary data model, and (b) the binary Gaussian mixture model. For both models, we derive game-theoretically optimal minimax privacy mechanisms, and show that the privacy mechanisms learned from data (in a generative adversarial fashion) match the theoretically optimal ones. This demonstrates that our framework can be easily applied in practice, even in the absence of dataset statistics.

cs.LG↗

arXiv2017

A solar blowout jet caused by the eruption of a magnetic flux rope

We investigate the three-dimensional (3D) magnetic structure of a blowout jet originated in the west edge of NOAA Active Region (AR) 11513 on 02 July 2012 by means of recently developed forced field extrapolation (FFE) model. The results show that the blowout jet was caused by the eruption of the magnetic flux rope (MFR) consisting of twisted field lines. We further calculate the twist number $\mathcal{T}_{w}$ and squashing factor Q of the reconstructed magnetic field and find that (1) the MFR corresponds well to the high $\mathcal{T}_{w}$ region (2) the MFR outer boundary corresponds well to the high Q region, probably interpreting the bright structure at the base of the jet. The twist number of the MFR is estimated to be $\mathcal{T}_{w}=-1.54\pm 0.67$. Thus, the kink instability is regarded as the initiation mechanism of the blowout jet as $\mathcal{T}_{w}$ reaching or even exceeding the threshold value of the kink instability. Our results also indicate that the bright point at the decaying phase is actually comprised of some small loops that are heated by the reconnection occurred above. In summary, the blowout jet is mostly consistent with the scenario proposed by \citet{mcs10} except that the kink instability is found to be a possible trigger.

astro-ph.SR↗

arXiv2017

Local Feature Descriptor Learning with Adaptive Siamese Network

Although the recent progress in the deep neural network has led to the development of learnable local feature descriptors, there is no explicit answer for estimation of the necessary size of a neural network. Specifically, the local feature is represented in a low dimensional space, so the neural network should have more compact structure. The small networks required for local feature descriptor learning may be sensitive to initial conditions and learning parameters and more likely to become trapped in local minima. In order to address the above problem, we introduce an adaptive pruning Siamese Architecture based on neuron activation to learn local feature descriptors, making the network more computationally efficient with an improved recognition rate over more complex networks. Our experiments demonstrate that our learned local feature descriptors outperform the state-of-art methods in patch matching.

cs.LG↗

arXiv2016

Market Segmentation for Privacy Differentiated "Free" Services

The emerging marketplace for online free services in which service providers earn revenue from using consumer data in direct and indirect ways has lead to significant privacy concerns. This leads to the following question: can the online marketplace sustain multiple service providers (SPs) that offer privacy-differentiated free services? This paper studies the problem of market segmentation for the free online services market by augmenting the classical Hotelling model for market segmentation analysis to include the fact that for the free services market, a consumer values service not in monetized terms but by its quality of service (QoS) and that the differentiator of services is not product price but the privacy risk advertised by a SP. Building upon the Hotelling model, this paper presents a parametrized model for SP profit and consumer valuation of service for both the two- and multi-SP problems to show that: (i) when consumers place a high value on privacy, it leads to a lower use of private data by SPs (i.e., their advertised privacy risk reduces), and thus, SPs compete on the QoS; (ii) SPs that are capable of differentiating on services that do not directly target consumers gain larger market share; and (iii) a higher valuation of privacy by consumers forces SPs with smaller untargeted revenue to offer lower privacy risk to attract more consumers. The work also illustrates the market segmentation problem for more than two SPs and highlights the instability of such markets.

cs.GT↗

arXiv2015

The role of active region coronal magnetic field in determining coronal mass ejection propagation direction

We study the role of the coronal magnetic field configuration of an active region in determining the propagation direction of a coronal mass ejection (CME). The CME occurred in the active region 11944 (S09W01) near the disk center on 2014 January 7 and was associated with an X1.2 flare. A new CME reconstruction procedure based on a polarimetric technique is adopted, which shows that the CME changed its propagation direction by around 28$^\circ$ in latitude within 2.5 R$_\odot$ and 43$^\circ$ in longitude within 6.5 R$_\odot$ with respect to the CME source region. This significant non-radial motion is consistent with the finding of M$\ddot{o}$stl et al. (2015). We use nonlinear force-free field (NLFFF) and potential field source surface (PFSS) extrapolation methods to determine the configurations of the coronal magnetic field. We also calculate the magnetic energy density distributions at different heights based on the extrapolations. Our results show that the active region coronal magnetic field has a strong influence on the CME propagation direction. This is consistent with the "channelling" by the active region coronal magnetic field itself, rather than deflection by nearby structures. These results indicate that the active region coronal magnetic field configuration has to be taken into account in order to determine CME propagation direction correctly.

astro-ph.SR↗

arXiv2015

Designing Incentive Schemes For Privacy-Sensitive Users

Businesses (retailers) often wish to offer personalized advertisements (coupons) to individuals (consumers), but run the risk of strong reactions from consumers who want a customized shopping experience but feel their privacy has been violated. Existing models for privacy such as differential privacy or information theory try to quantify privacy risk but do not capture the subjective experience and heterogeneous expression of privacy-sensitivity. We propose a Markov decision process (MDP) model to capture (i) different consumer privacy sensitivities via a time-varying state; (ii) different coupon types (action set) for the retailer; and (iii) the action-and-state-dependent cost for perceived privacy violations. For the simple case with two states ("Normal" and "Alerted"), two coupons (targeted and untargeted) model, and consumer behavior statistics known to the retailer, we show that a stationary threshold-based policy is the optimal coupon-offering strategy for a retailer that wishes to minimize its expected discounted cost. The threshold is a function of all model parameters; the retailer offers a targeted coupon if their belief that the consumer is in the "Alerted" state is below the threshold. We extend this two-state model to consumers with multiple privacy-sensitivity states as well as coupon-dependent state transition probabilities. Furthermore, we study the case with imperfect (noisy) cost feedback from consumers and uncertain initial belief state.

cs.GT↗

arXiv2012

The Morphologic Properties of Magnetic networks over the Solar Cycle 23

The morphologic properties of the magnetic networks during Carrington Rotations (CR) 1955 to 2091 (from 1999 to 2010) have been analyzed by applying the watershed algorithm to magnetograms observed by the Michelson Doppler Interferometer (MDI) on board the Solar and Heliospheric Observatory (SOHO) spacecraft. We find that the average area of magnetic cells on the solar surface at lower latitudes (within +-50 degree) are smaller than those at higher latitudes (beyond +-50 degree). Statistical analysis of these data indicates that the magnetic networks are of fractal in nature, and the average fractal dimension is D_f = 1.253+-0.011. We also find that both the fractal dimension and the size of the magnetic networks are anti-correlated with the sunspot area. This is perhaps because a strong magnetic field can suppress spatially modulated oscillation, compress the boundaries of network cells, leading to smoother cell boundaries. The fractal dimension of the cell deviates that predicted from an isobar of Kolmogorov homogeneous turbulence.

astro-ph.SR↗