The adversarial machine learning literature is largely partitioned into evasion attacks on testing data and poisoning attacks on training data. In this work, we show that adversarial examples, originally intended for attacking pre-trained models, are even more effective for data poisoning than recent methods designed specifically for poisoning. In fact, adversarial examples with labels re-assigned by the crafting network remain effective for training, suggesting that adversarial examples contain useful semantic content, just with the "wrong" labels (according to a network, but not a human). Our method, adversarial poisoning, is substantially more effective than existing poisoning methods for secure dataset release, and we release a poisoned version of ImageNet, ImageNet-P, to encourage research into the strength of this form of data obfuscation.

We propose a new algorithm to incorporate class conditional information into the critic of GANs via a multi-class generalization of the commonly used Hinge loss that is compatible with both supervised and semi-supervised settings. We study the compromise between training a state of the art generator and an accurate classifier simultaneously, and propose a way to use our algorithm to measure the degree to which a generator and critic are class conditional. We show the trade-off between a generator-critic pair respecting class conditioning inputs and generating the highest quality images. With our multi-hinge loss modification we are able to improve Inception Scores and Frechet Inception Distance on the Imagenet dataset.

Recent developments in machine learning and signal processing have resulted in many new techniques that are able to effectively capture the intrinsic yet complex properties of hyperspectral imagery (HSI). Tasks ranging from anomaly detection to classification can now be solved by taking advantage of very efficient algorithms which have their roots in representation theory and computational approximation. Time–frequency methods are one example of such techniques. They provide means to analyze and extract the spectral content from data. On the other hand, hierarchical methods such as neural networks (NNs) incorporate spatial information across scales and model multiple levels of dependencies between spectral features. Both of these approaches have recently been proven to provide significant advances in the spectral-spatial classification of HSI. The 3-D Fourier scattering transform, which is introduced in this article, is an amalgamation of time–frequency representations with NN architectures. It leverages the benefits provided by the short-time Fourier transform with the numerical efficiency of deep learning network structures. We test the proposed method on several standard hyperspectral data sets, and we present results that indicate that the 3-D Fourier scattering transform is highly effective at representing spectral content when compared with other state-of-the-art spectral-spatial classification methods.

We present a new physics-inspired method for analysis of hyperspectral imagery (HSI). The method is based on the concept of transport models for graphs. The proposed approach generalizes existing dimension reduction and feature extraction algorithms, by replacing the role of diffusion processes, as a measure of estimating proximity, with dynamical systems. This approach allows us to exploit different and new relationships within the complex data structures, such as those arising in HSI. We demonstrate this by proposing a specific multi-scale algorithm in which transport models are used to translate the information about contextual similarities of material classes to enhance feature extraction and classification results. This point is illustrated with a series of computational experiments.

This paper deals with tensor completion for the solution of multidimensional inverse problems arising in nuclear magnetic resonance (NMR) relaxometry. We study the problem of reconstructing an approximately low-rank tensor from a small number of noisy linear measurements. New recovery guarantees, numerical algorithms, nonuniform sampling strategies, and parameter selection methods are developed in this context. In particular, we derive a fixed point continuation algorithm for tensor completion and prove its convergence. A restricted isometry property-based tensor recovery guarantee is proved. Probabilistic recovery guarantees are obtained for sub-Gaussian measurement operators and for measurements obtained by nonuniform sampling from a Parseval tight frame. The proposed algorithm is then applied to the setting of nuclear magnetic resonance relaxometry, for both simulated and experimental data. We compare our results with basis pursuit as well as with the state-of-the-art nonsubsampled data acquisition and reconstruction approach. Our experiments indicate that tensor recovery promises to significantly accelerate $N$-dimensional NMR relaxometry and related experiments, enabling previously impractical experiments to be performed. Our methods could also be applied to other similar inverse problems arising in machine learning, signal and image processing, and computer vision.

This paper considers attacks against machine learning algorithms used in remote sensing applications. The remote sensing domain presents a suite of challenges that are not fully addressed by current research focused on natural image data. In this paper we present a new study of adversarial examples in the context of satellite image classification problems. Using a recently curated data set and associated classifier, we provide a preliminary analysis of adversarial examples in settings where the targeted classifier is permitted multiple observations of the same location over time. While our experiments to date are purely digital, our problem setup incorporates a number of practical considerations that an attacker would need to take into account when mounting physical attacks.

We develop an algorithm for single-image superresolution of remotely sensed data based on the discrete shearlet transform. The shearlet transform extracts directional features of signals and is known to provide near-optimally sparse representations for a broad class of images. This often leads to superior performance in edge detection and image representation when compared with isotropic frames. We justify the use of shearlets mathematically, before presenting a denoising single-image superresolution algorithm that combines the shearlet transform with sparse mixing estimators (SMEs). Our algorithm is compared with a variety of single-image superresolution methods, including wavelet SME superresolution. Our numerical results demonstrate competitive performance in terms of peak-signal-to-noise ratio and structural similarity index metric.

In this paper we propose to utilize a new concept of discrete directional Gabor frames for automatic image registration. The directional Gabor representations have been shown to provide more accurate feature extraction than directional wavelet transforms for images where texture is the dominant feature. Initial experimental results are presented here which indicate that discrete directional Gabor frames exhibit strong correlations, which indicates that they are likely to improve the existing image registration toolbox.

As new sensing modalities emerge and the presence of multiple sensors per platform becomes widespread, it is vital to develop new algorithms and techniques which can fuse this data. Previous attempts to deal with the problem of heterogeneous data integration for the applications in data classification were either highly data dependent or relied on simply fusing classifier outputs. In this paper we examine several related approaches: graph fusion, operator fusion, and feature space fusion. They are all associated with graph diffusion processes generated by appropriately designed operators. Our results do not make any assumptions about the data and can be easily extended to new additional modalities.

We present an algorithm to solve the two-dimensional Fredholm integral of the first kind with tensor product structure from a limited number of measurements, with the goal of using this method to speed up nuclear magnetic resonance spectroscopy. This is done by incorporating compressive sensing-type arguments to fill in missing measurements, using a priori knowledge of the structure of the data. In the first step we recover a compressed data matrix from measurements that form a tight frame, and establish that these measurements satisfy the restricted isometry property. Recovery can be done from as few as 10% of the total measurements. In the second and third steps, we solve the zeroth-order regularization minimization problem using the Venkataramanan-Song-Hürlimann algorithm. We demonstrate the performance of this algorithm on simulated data and show that our approach is a realistic approach to speeding up the data acquisition.

We introduce Schroedinger Eigenmaps (SE), a new semi-supervised manifold learning and recovery technique. This method is based on an implementation of graph Schroedinger operators with appropriately constructed barrier potentials as carriers of labeled information. We use our approach for the analysis of standard biomedical datasets and new multispectral retinal images.

We propose a new kernel and frame based dimension reducing algorithm by exploiting the synergy between endmembers and kernel based classification schemes. This synergy becomes available by means of utilizing the notions of frames and frame representations.