More Like this

1. Graph interpolating activation improves both natural and robust accuracies in data-efficient deep learning

   https://doi.org/10.1017/S0956792520000406  

   WANG, BAO; OSHER, STAN J. (January 2020, European Journal of Applied Mathematics)

   null (Ed.)

   Improving the accuracy and robustness of deep neural nets (DNNs) and adapting them to small training data are primary tasks in deep learning (DL) research. In this paper, we replace the output activation function of DNNs, typically the data-agnostic softmax function, with a graph Laplacian-based high-dimensional interpolating function which, in the continuum limit, converges to the solution of a Laplace–Beltrami equation on a high-dimensional manifold. Furthermore, we propose end-to-end training and testing algorithms for this new architecture. The proposed DNN with graph interpolating activation integrates the advantages of both deep learning and manifold learning. Compared to the conventional DNNs with the softmax function as output activation, the new framework demonstrates the following major advantages: First, it is better applicable to data-efficient learning in which we train high capacity DNNs without using a large number of training data. Second, it remarkably improves both natural accuracy on the clean images and robust accuracy on the adversarial images crafted by both white-box and black-box adversarial attacks. Third, it is a natural choice for semi-supervised learning. This paper is a significant extension of our earlier work published in NeurIPS, 2018. For reproducibility, the code is available at https://github.com/BaoWangMath/DNN-DataDependentActivation . 

   more » « less
2. Comparing Detection Schemes for Adversarial Images against Deep Learning Models for Cancer Imaging

   https://doi.org/10.3390/cancers15051548  

   Joel, Marina Z.; Avesta, Arman; Yang, Daniel X.; Zhou, Jian-Ge; Omuro, Antonio; Herbst, Roy S.; Krumholz, Harlan M.; Aneja, Sanjay (March 2023, Cancers)

   Deep learning (DL) models have demonstrated state-of-the-art performance in the classification of diagnostic imaging in oncology. However, DL models for medical images can be compromised by adversarial images, where pixel values of input images are manipulated to deceive the DL model. To address this limitation, our study investigates the detectability of adversarial images in oncology using multiple detection schemes. Experiments were conducted on thoracic computed tomography (CT) scans, mammography, and brain magnetic resonance imaging (MRI). For each dataset we trained a convolutional neural network to classify the presence or absence of malignancy. We trained five DL and machine learning (ML)-based detection models and tested their performance in detecting adversarial images. Adversarial images generated using projected gradient descent (PGD) with a perturbation size of 0.004 were detected by the ResNet detection model with an accuracy of 100% for CT, 100% for mammogram, and 90.0% for MRI. Overall, adversarial images were detected with high accuracy in settings where adversarial perturbation was above set thresholds. Adversarial detection should be considered alongside adversarial training as a defense technique to protect DL models for cancer imaging classification from the threat of adversarial images. 

   more » « less
3. Robust Learning via Ensemble Density Propagation in Deep Neural Networks

   https://doi.org/10.1109/MLSP49062.2020.9231635  

   Carannante, Giuseppina; Dera, Dimah; Rasool, Ghulam; Bouaynaya, Nidhal C.; Mihaylova, Lyudmila (September 2020, IEEE 30th International Workshop on Machine Learning for Signal Processing (MLSP))

   null (Ed.)

   Learning in uncertain, noisy, or adversarial environments is a challenging task for deep neural networks (DNNs). We propose a new theoretically grounded and efficient approach for robust learning that builds upon Bayesian estimation and Variational Inference. We formulate the problem of density propagation through layers of a DNN and solve it using an Ensemble Density Propagation (EnDP) scheme. The EnDP approach allows us to propagate moments of the variational probability distribution across the layers of a Bayesian DNN, enabling the estimation of the mean and covariance of the predictive distribution at the output of the model. Our experiments using MNIST and CIFAR-10 datasets show a significant improvement in the robustness of the trained models to random noise and adversarial attacks. 

   more » « less
4. Revisiting Non-Parametric Matching Cost Volumes for Robust and Generalizable Stereo Matching

   Cheng, Kelvin; Wu, Tianfu; Healey, Christopher G. (December 2022, Advances in Neural Information Processing Systems)

   The integration of DNN-contextualized binary-pattern-driven non-parametric cost volume and DNN cost aggregation leads to more robust and more generalizable stereo matching. Abstract: Stereo matching is a classic challenging problem in computer vision, which has recently witnessed remarkable progress by Deep Neural Networks (DNNs). This paradigm shift leads to two interesting and entangled questions that have not been addressed well. First, it is unclear whether stereo matching DNNs that are trained from scratch really learn to perform matching well. This paper studies this problem from the lens of white-box adversarial attacks. It presents a method of learning stereo-constrained photometrically-consistent attacks, which by design are weaker adversarial attacks, and yet can cause catastrophic performance drop for those DNNs. This observation suggests that they may not actually learn to perform matching well in the sense that they should otherwise achieve potentially even better after stereo-constrained perturbations are introduced. Second, stereo matching DNNs are typically trained under the simulation-to-real (Sim2Real) pipeline due to the data hungriness of DNNs. Thus, alleviating the impacts of the Sim2Real photometric gap in stereo matching DNNs becomes a pressing need. Towards joint adversarially robust and domain generalizable stereo matching, this paper proposes to learn DNN-contextualized binary-pattern-driven non-parametric cost-volumes. It leverages the perspective of learning the cost aggregation via DNNs, and presents a simple yet expressive design that is fully end-to-end trainable, without resorting to specific aggregation inductive biases. In experiments, the proposed method is tested in the SceneFlow dataset, the KITTI2015 dataset, and the Middlebury dataset. It significantly improves the adversarial robustness, while retaining accuracy performance comparable to state-of-the-art methods. It also shows a better Sim2Real generalizability. Our code and pretrained models are released at \href{https://github.com/kelkelcheng/AdversariallyRobustStereo}{this Github Repo}. 

   more » « less
5. Image-based PDF Malware Detection Using Pre-trained Deep Neural Networks

   https://doi.org/10.1109/ISDFS60797.2024.10527343  

   Nichols, Tyler; Zemlanicky, Jack; Luo, Zhirui; Li, Qingqing; Zheng, Jun (April 2024, 2024 12th International Symposium on Digital Forensics and Security (ISDFS))

   PDF is a popular document file format with a flexible file structure that can embed diverse types of content, including images and JavaScript code. However, these features make it a favored vehicle for malware attackers. In this paper, we propose an image-based PDF malware detection method that utilizes pre-trained deep neural networks (DNNs). Specifically, we convert PDF files into fixed-size grayscale images using an image visualization technique. These images are then fed into pre-trained DNN models to classify them as benign or malicious. We investigated four classical pre-trained DNN models in our study. We evaluated the performance of the proposed method using the publicly available Contagio PDF malware dataset. Our results demonstrate that MobileNetv3 achieves the best detection performance with an accuracy of 0.9969 and exhibits low computational complexity, making it a promising solution for image-based PDF malware detection. 

   more » « less