skip to main content
Explore Scholarly Publications and Datasets in the NSF-PAR

  • Home
  • Search Results
  • Page 1 of 49

Search for: All records

Creators/Authors contains: "Zhang, H."

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. ; ; ; ; ( , NeurIPS)
    Free, publicly-accessible full text available December 1, 2022
  2. ; ( , IJCAI)
    Recent work on explainable clustering allows describing clusters when the features are interpretable. However, much modern machine learning focuses on complex data such as images, text, and graphs where deep learning is used but the raw features of data are not interpretable. This paper explores a novel setting for performing clustering on complex data while simultaneously generating explanations using interpretable tags. We propose deep descriptive clustering that performs sub-symbolic representation learning on complex data while generating explanations based on symbolic data. We form good clusters by maximizing the mutual information between empirical distribution on the inputs and the induced clusteringmore »labels for clustering objectives. We generate explanations by solving an integer linear programming that generates concise and orthogonal descriptions for each cluster. Finally, we allow the explanation to inform better clustering by proposing a novel pairwise loss with self-generated constraints to maximize the clustering and explanation module's consistency. Experimental results on public data demonstrate that our model outperforms competitive baselines in clustering performance while offering high-quality cluster-level explanations.« less
    Free, publicly-accessible full text available August 1, 2022
  3. ; ; ; ( , 30th USENIX Security Symposium (USENIX Security 21))
    With their growing popularity, Internet-of-Things (IoT) devices have become attractive targets for attack. Like most modern software systems, IoT device firmware depends on external third-party libraries extensively, increasing the attack surface of IoT devices. Furthermore, we find that the risk is compounded by inconsistent library management practices and delays in applying security updates—sometimes hundreds of days behind the public availability of critical patches—by device vendors. Worse yet, because these dependencies are "baked into" the vendor-controlled firmware, even security-conscious users are unable to take matters into their own hands when it comes to good security hygiene. We present Capture, a novelmore »architecture for deploying IoT device firmware that addresses this problem by allowing devices on a local network to leverage a centralized hub with third-party libraries that are managed and kept up-to-date by a single trusted entity. An IoT device supporting Capture comprises of two components: Capture-enabled firmware on the device and a remote driver that uses third-party libraries on the Capture hub in the local network. To ensure isolation, we introduce a novel Virtual Device Entity (VDE) interface that facilitates access control between mutually-distrustful devices that reside on the same hub. Our evaluation on a prototype implementation of Capture, along with 9 devices and 3 automation applets ported to our framework, shows that our approach incurs low overhead in most cases (<15% increased latency, <10% additional resources). We show that a single Capture Hub with modest hardware can support hundreds of devices, keeping their shared libraries up-to-date.« less
    Free, publicly-accessible full text available August 1, 2022
  4. ; ; ; ; ; ; ; ; ; ; et al ( , Workshop on Hyperspectral Image and Signal Processing Evolution in Remote Sensing)
    Multi- and hyperspectral imaging modalities encompass a growing number of spectral techniques that find many applications in geospatial, biomedical and machine vision fields. The rapidly increasing number of applications requires a convenient easy-to-navigate software that can be used by new and experienced users to analyze data, develop, apply, and deploy novel algorithms. Herein, we present our platform, IDCube that performs essential operations in hyperspectral data analysis to realize the full potential of spectral imaging. The strength of the software lies in its interactive features that enable the users to optimize parameters and obtain visual input for the user. The entiremore »software can be operated without any prior programming skills allowing interactive sessions of raw and processed data. IDCube Lite, a free version of the software described in the paper, has many benefits compared to existing packages and offers structural flexibility to discover new hidden features.« less
    Free, publicly-accessible full text available July 19, 2022
  5. ; ; ; ; ; ( , Advances in neural information processing systems)
    Accepted Manuscript Full Text Available
  6. ; ; ; ; ; ( , Advances in neural information processing systems)
    Accepted Manuscript Full Text Available
  7. ; ; ; ; ; ; ; ( , Earth's Future)
    Full Text Available 
  8. ; ; ; ; ; ; ; ( , Physical Review Letters)
    Full Text Available 
  9. ; ; ; ; ; ( , Dental Materials)
    Full Text Available 
  10. ; ; ; ; ; ; ( , Proc. of 18th IEEE International Conference on Dependable, Autonomic and Secure Computing (DASC))
    Denial of Service (DoS) is one of the common attempts in security hacking for making computation resources unavailable or to impair geographical networks. In this paper, we detect Denial of Service (DoS) attack from publicly available datasets using Logistic regression, Naive Bayes algorithm and artificial neural networks. The results from our experiments indicate that the accuracy, ROC curve and balanced accuracy of artificial neural network were higher than Naive Bayes algorithm and logistic regression for slightly imbalanced distribution dataset.
    Accepted Manuscript Full Text Available