arXiv subjects

Cong Zhao

Publications and source records attributed to Cong Zhao.

33 records · Page 2Linked to original sources

arXiv2023

Weakly-Supervised Action Localization by Hierarchically-structured Latent Attention Modeling

Weakly-supervised action localization aims to recognize and localize action instancese in untrimmed videos with only video-level labels. Most existing models rely on multiple instance learning(MIL), where the predictions of unlabeled instances are supervised by classifying labeled bags. The MIL-based methods are relatively well studied with cogent performance achieved on classification but not on localization. Generally, they locate temporal regions by the video-level classification but overlook the temporal variations of feature semantics. To address this problem, we propose a novel attention-based hierarchically-structured latent model to learn the temporal variations of feature semantics. Specifically, our model entails two components, the first is an unsupervised change-points detection module that detects change-points by learning the latent representations of video features in a temporal hierarchy based on their rates of change, and the second is an attention-based classification model that selects the change-points of the foreground as the boundaries. To evaluate the effectiveness of our model, we conduct extensive experiments on two benchmark datasets, THUMOS-14 and ActivityNet-v1.3. The experiments show that our method outperforms current state-of-the-art methods, and even achieves comparable performance with fully-supervised methods.

cs.CV↗

arXiv2023

BGO quenching effect on spectral measurements of cosmic-ray nuclei in DAMPE experiment

The Dark Matter Particle Explorer (DAMPE) is a satellite-borne detector designed to measure high energy cosmic-rays and $γ$-rays. As a key sub-detector of DAMPE, the Bismuth Germanium Oxide (BGO) imaging calorimeter is utilized to measure the particle energy with a high resolution. The nonlinear fluorescence response of BGO for large ionization energy deposition, known as the quenching effect, results in an under-estimate of the energy measurement for cosmic-ray nuclei. In this paper, various models are employed to characterize the BGO quenching factors obtained from the experimental data of DAMPE. Applying the proper quenching model in the detector simulation process, we investigate the tuned energy responses for various nuclei and compare the results based on two different simulation softwares, i.e. GEANT4 and FLUKA. The BGO quenching effect results in a decrease of the measured energy by approximately $2.5\%$ ($5.7 \%$) for carbon (iron) at $\sim$10 GeV/n and $<1\%$ above 1 TeV/n, respectively. Accordingly, the correction of the BGO quenching effect leads to an increase of the low-energy flux measurement of cosmic-ray nuclei.

hep-ex↗

arXiv2022

LDP-IDS: Local Differential Privacy for Infinite Data Streams

Streaming data collection is essential to real-time data analytics in various IoTs and mobile device-based systems, which, however, may expose end users' privacy. Local differential privacy (LDP) is a promising solution to privacy-preserving data collection and analysis. However, existing few LDP studies over streams are either applicable to finite streams only or suffering from insufficient protection. This paper investigates this problem by proposing LDP-IDS, a novel $w$-event LDP paradigm to provide practical privacy guarantee for infinite streams at users end, and adapting the popular budget division framework in centralized differential privacy (CDP). By constructing a unified error analysi for LDP, we first develop two adatpive budget division-based LDP methods for LDP-IDS that can enhance data utility via leveraging the non-deterministic sparsity in streams. Beyond that, we further propose a novel population division framework that can not only avoid the high sensitivity of LDP noise to budget division but also require significantly less communication. Based on the framework, we also present two adaptive population division methods for LDP-IDS with theoretical analysis. We conduct extensive experiments on synthetic and real-world datasets to evaluate the effectiveness and efficiency pf our proposed frameworks and methods. Experimental results demonstrate that, despite the effectiveness of the adaptive budget division methods, the proposed population division framework and methods can further achieve much higher effectiveness and efficiency.

cs.DB↗

arXiv2022

ACE: Towards Application-Centric Edge-Cloud Collaborative Intelligence

Intelligent applications based on machine learning are impacting many parts of our lives. They are required to operate under rigorous practical constraints in terms of service latency, network bandwidth overheads, and also privacy. Yet current implementations running in the Cloud are unable to satisfy all these constraints. The Edge-Cloud Collaborative Intelligence (ECCI) paradigm has become a popular approach to address such issues, and rapidly increasing applications are developed and deployed. However, these prototypical implementations are developer-dependent and scenario-specific without generality, which cannot be efficiently applied in large-scale or to general ECC scenarios in practice, due to the lack of supports for infrastructure management, edge-cloud collaborative service, complex intelligence workload, and efficient performance optimization. In this article, we systematically design and construct the first unified platform, ACE, that handles ever-increasing edge and cloud resources, user-transparent services, and proliferating intelligence workloads with increasing scale and complexity, to facilitate cost-efficient and high-performing ECCI application development and deployment. For verification, we explicitly present the construction process of an ACE-based intelligent video query application, and demonstrate how to achieve customizable performance optimization efficiently. Based on our initial experience, we discuss both the limitations and vision of ACE to shed light on promising issues to elaborate in the approaching ECCI ecosystem.

cs.DC↗

arXiv2021

IRONWAN: Increasing Reliability of Overlapping Networks in LoRaWAN

LoRaWAN deployments follow an ad-hoc deployment model that has organically led to overlapping communication networks, sharing the wireless spectrum, and completely unaware of each other. LoRaWAN uses ALOHA-style communication where it is almost impossible to schedule transmission between networks belonging to different owners properly. The inability to schedule overlapping networks will cause inter-network interference, which will increase node-to-gateway message losses and gateway-to-node acknowledgement failures. This problem is likely to get worse as the number of LoRaWAN networks increase. In response to this problem, we propose IRONWAN, a wireless overlay network that shares communication resources without modifications to underlying protocols. It utilises the broadcast nature of radio communication and enables gateway-to-gateway communication to facilitate the search for failed messages and transmit failed acknowledgements already received and cached in overlapping network's gateways. IRONWAN uses two novel algorithms, a Real-time Message Inter-arrival Predictor, to highlight when a server has not received an expected uplink message. The Interference Predictor ensures that extra gateway-to-gateway communication does not negatively impact communication bandwidth. We evaluate IRONWAN on a 1000-node simulator with up to ten gateways and a 10-node testbed with 2-gateways. Results show that IRONWAN can achieve up to 12\% higher packet delivery ratio (PDR) and total messages received per node while increasing the minimum PDR by up to 28\%. These improvements save up to 50\% node's energy. Finally, we demonstrate that IRONWAN has comparable performance to an optimal solution (wired, centralised) but with 2-32 times lower communication costs. IRONWAN also has up to 14\% better PDR when compared to FLIP, a wired-distributed gateway-to-gateway protocol in certain scenarios.

cs.NI↗

arXiv2020

HEU Emotion: A Large-scale Database for Multi-modal Emotion Recognition in the Wild

The study of affective computing in the wild setting is underpinned by databases. Existing multimodal emotion databases in the real-world conditions are few and small, with a limited number of subjects and expressed in a single language. To meet this requirement, we collected, annotated, and prepared to release a new natural state video database (called HEU Emotion). HEU Emotion contains a total of 19,004 video clips, which is divided into two parts according to the data source. The first part contains videos downloaded from Tumblr, Google, and Giphy, including 10 emotions and two modalities (facial expression and body posture). The second part includes corpus taken manually from movies, TV series, and variety shows, consisting of 10 emotions and three modalities (facial expression, body posture, and emotional speech). HEU Emotion is by far the most extensive multi-modal emotional database with 9,951 subjects. In order to provide a benchmark for emotion recognition, we used many conventional machine learning and deep learning methods to evaluate HEU Emotion. We proposed a Multi-modal Attention module to fuse multi-modal features adaptively. After multi-modal fusion, the recognition accuracies for the two parts increased by 2.19% and 4.01% respectively over those of single-modal facial expression recognition.

cs.CV↗

arXiv2020

CDC: Classification Driven Compression for Bandwidth Efficient Edge-Cloud Collaborative Deep Learning

The emerging edge-cloud collaborative Deep Learning (DL) paradigm aims at improving the performance of practical DL implementations in terms of cloud bandwidth consumption, response latency, and data privacy preservation. Focusing on bandwidth efficient edge-cloud collaborative training of DNN-based classifiers, we present CDC, a Classification Driven Compression framework that reduces bandwidth consumption while preserving classification accuracy of edge-cloud collaborative DL. Specifically, to reduce bandwidth consumption, for resource-limited edge servers, we develop a lightweight autoencoder with a classification guidance for compression with classification driven feature preservation, which allows edges to only upload the latent code of raw data for accurate global training on the Cloud. Additionally, we design an adjustable quantization scheme adaptively pursuing the tradeoff between bandwidth consumption and classification accuracy under different network conditions, where only fine-tuning is required for rapid compression ratio adjustment. Results of extensive experiments demonstrate that, compared with DNN training with raw data, CDC consumes 14.9 times less bandwidth with an accuracy loss no more than 1.06%, and compared with DNN training with data compressed by AE without guidance, CDC introduces at least 100% lower accuracy loss.

cs.LG↗

arXiv2020

OL4EL: Online Learning for Edge-cloud Collaborative Learning on Heterogeneous Edges with Resource Constraints

Distributed machine learning (ML) at network edge is a promising paradigm that can preserve both network bandwidth and privacy of data providers. However, heterogeneous and limited computation and communication resources on edge servers (or edges) pose great challenges on distributed ML and formulate a new paradigm of Edge Learning (i.e. edge-cloud collaborative machine learning). In this article, we propose a novel framework of 'learning to learn' for effective Edge Learning (EL) on heterogeneous edges with resource constraints. We first model the dynamic determination of collaboration strategy (i.e. the allocation of local iterations at edge servers and global aggregations on the Cloud during collaborative learning process) as an online optimization problem to achieve the tradeoff between the performance of EL and the resource consumption of edge servers. Then, we propose an Online Learning for EL (OL4EL) framework based on the budget-limited multi-armed bandit model. OL4EL supports both synchronous and asynchronous learning patterns, and can be used for both supervised and unsupervised learning tasks. To evaluate the performance of OL4EL, we conducted both real-world testbed experiments and extensive simulations based on docker containers, where both Support Vector Machine and K-means were considered as use cases. Experimental results demonstrate that OL4EL significantly outperforms state-of-the-art EL and other collaborative ML approaches in terms of the trade-off between learning performance and resource consumption.

cs.DC↗

arXiv2020

SurveilEdge: Real-time Video Query based on Collaborative Cloud-Edge Deep Learning

The real-time query of massive surveillance video data plays a fundamental role in various smart urban applications such as public safety and intelligent transportation. Traditional cloud-based approaches are not applicable because of high transmission latency and prohibitive bandwidth cost, while edge devices are often incapable of executing complex vision algorithms with low latency and high accuracy due to restricted resources. Given the infeasibility of both cloud-only and edge-only solutions, we present SurveilEdge, a collaborative cloud-edge system for real-time queries of large-scale surveillance video streams. Specifically, we design a convolutional neural network (CNN) training scheme to reduce the training time with high accuracy, and an intelligent task allocator to balance the load among different computing nodes and to achieve the latency-accuracy tradeoff for real-time queries. We implement SurveilEdge on a prototype with multiple edge devices and a public Cloud, and conduct extensive experiments using realworld surveillance video datasets. Evaluation results demonstrate that SurveilEdge manages to achieve up to 7x less bandwidth cost and 5.4x faster query response time than the cloud-only solution; and can improve query accuracy by up to 43.9% and achieve 15.8x speedup respectively, in comparison with edge-only approaches.

cs.DC↗

arXiv2019

Asynchronous Federated Learning with Differential Privacy for Edge Intelligence

Federated learning has been showing as a promising approach in paving the last mile of artificial intelligence, due to its great potential of solving the data isolation problem in large scale machine learning. Particularly, with consideration of the heterogeneity in practical edge computing systems, asynchronous edge-cloud collaboration based federated learning can further improve the learning efficiency by significantly reducing the straggler effect. Despite no raw data sharing, the open architecture and extensive collaborations of asynchronous federated learning (AFL) still give some malicious participants great opportunities to infer other parties' training data, thus leading to serious concerns of privacy. To achieve a rigorous privacy guarantee with high utility, we investigate to secure asynchronous edge-cloud collaborative federated learning with differential privacy, focusing on the impacts of differential privacy on model convergence of AFL. Formally, we give the first analysis on the model convergence of AFL under DP and propose a multi-stage adjustable private algorithm (MAPA) to improve the trade-off between model utility and privacy by dynamically adjusting both the noise scale and the learning rate. Through extensive simulations and real-world experiments with an edge-could testbed, we demonstrate that MAPA significantly improves both the model accuracy and convergence speed with sufficient privacy guarantee.

cs.LG↗

arXiv2019

Neural Architecture Refinement: A Practical Way for Avoiding Overfitting in NAS

Neural architecture search (NAS) is proposed to automate the architecture design process and attracts overwhelming interest from both academia and industry. However, it is confronted with overfitting issue due to the high-dimensional search space composed by operator selection and skip connection of each layer. This paper explores the architecture overfitting issue in depth based on the reinforcement learning-based NAS framework. We show that the policy gradient method has deep correlations with the cross entropy minimization. Based on this correlation, we further demonstrate that, though the reward of NAS is sparse, the policy gradient method implicitly assign the reward to all operations and skip connections based on the sampling frequency. However, due to the inaccurate reward estimation, curse of dimensionality problem and the hierachical structure of neural networks, reward charateristics for operators and skip connections have intrinsic differences, the assigned rewards for the skip connections are extremely noisy and inaccurate. To alleviate this problem, we propose a neural architecture refinement approach that working with an initial state-of-the-art network structure and only refining its operators. Extensive experiments have demonstrated that the proposed method can achieve fascinated results, including classification, face recognition etc.

cs.LG↗

arXiv2018

Coarse-to-Fine Annotation Enrichment for Semantic Segmentation Learning

Rich high-quality annotated data is critical for semantic segmentation learning, yet acquiring dense and pixel-wise ground-truth is both labor- and time-consuming. Coarse annotations (e.g., scribbles, coarse polygons) offer an economical alternative, with which training phase could hardly generate satisfactory performance unfortunately. In order to generate high-quality annotated data with a low time cost for accurate segmentation, in this paper, we propose a novel annotation enrichment strategy, which expands existing coarse annotations of training data to a finer scale. Extensive experiments on the Cityscapes and PASCAL VOC 2012 benchmarks have shown that the neural networks trained with the enriched annotations from our framework yield a significant improvement over that trained with the original coarse labels. It is highly competitive to the performance obtained by using human annotated dense annotations. The proposed method also outperforms among other state-of-the-art weakly-supervised segmentation methods.

cs.CV↗

arXiv2017

Towards Accurate Binary Convolutional Neural Network

We introduce a novel scheme to train binary convolutional neural networks (CNNs) -- CNNs with weights and activations constrained to {-1,+1} at run-time. It has been known that using binary weights and activations drastically reduce memory size and accesses, and can replace arithmetic operations with more efficient bitwise operations, leading to much faster test-time inference and lower power consumption. However, previous works on binarizing CNNs usually result in severe prediction accuracy degradation. In this paper, we address this issue with two major innovations: (1) approximating full-precision weights with the linear combination of multiple binary weight bases; (2) employing multiple binary activations to alleviate information loss. The implementation of the resulting binary CNN, denoted as ABC-Net, is shown to achieve much closer performance to its full-precision counterpart, and even reach the comparable prediction accuracy on ImageNet and forest trail datasets, given adequate binary weight bases and activations.

cs.LG↗

arXiv2017

Rapid, User-Transparent, and Trustworthy Device Pairing for D2D-Enabled Mobile Crowdsourcing

Mobile Crowdsourcing is a promising service paradigm utilizing ubiquitous mobile devices to facilitate largescale crowdsourcing tasks (e.g. urban sensing and collaborative computing). Many applications in this domain require Device-to-Device (D2D) communications between participating devices for interactive operations such as task collaborations and file transmissions. Considering the private participating devices and their opportunistic encountering behaviors, it is highly desired to establish secure and trustworthy D2D connections in a fast and autonomous way, which is vital for implementing practical Mobile Crowdsourcing Systems (MCSs). In this paper, we develop an efficient scheme, Trustworthy Device Pairing (TDP), which achieves user-transparent secure D2D connections and reliable peer device selections for trustworthy D2D communications. Through rigorous analysis, we demonstrate the effectiveness and security intensity of TDP in theory. The performance of TDP is evaluated based on both real-world prototype experiments and extensive trace-driven simulations. Evaluation results verify our theoretical analysis and show that TDP significantly outperforms existing approaches in terms of pairing speed, stability, and security.

cs.NI↗

arXiv2017

Cheating-Resilient Incentive Scheme for Mobile Crowdsensing Systems

Mobile Crowdsensing is a promising paradigm for ubiquitous sensing, which explores the tremendous data collected by mobile smart devices with prominent spatial-temporal coverage. As a fundamental property of Mobile Crowdsensing Systems, temporally recruited mobile users can provide agile, fine-grained, and economical sensing labors, however their self-interest cannot guarantee the quality of the sensing data, even when there is a fair return. Therefore, a mechanism is required for the system server to recruit well-behaving users for credible sensing, and to stimulate and reward more contributive users based on sensing truth discovery to further increase credible reporting. In this paper, we develop a novel Cheating-Resilient Incentive (CRI) scheme for Mobile Crowdsensing Systems, which achieves credibility-driven user recruitment and payback maximization for honest users with quality data. Via theoretical analysis, we demonstrate the correctness of our design. The performance of our scheme is evaluated based on extensive realworld trace-driven simulations. Our evaluation results show that our scheme is proven to be effective in terms of both guaranteeing sensing accuracy and resisting potential cheating behaviors, as demonstrated in practical scenarios, as well as those that are intentionally harsher.

cs.NI↗