Evasion attacks in adversarial machine learning (AML) have been well-established for multi-class classification problems. However, for multi-label audio classification, several considerations and unique challenges exist in evasion AML that the attacker needs to consider. First, the attacker must cater the attack from the multi-class scenario assumed by the vast majority of evasion attacks to multi-label AML. Second, the attacker should consider creating imper-ceptible adversarial examples to reduce suspicion of the human element by factoring in properties inherent to the audio data set and model. Lastly, with the proliferation of AML techniques, the attacker should assume that the target model will implement a deep neural network (DNN) defense and should attempt a strategy to overcome those defenses. In this paper, we show how an attacker would cater and design an attack against multi-label audio classification using stochastic DNNs, highlighting some considerations for linking AML techniques to acoustic-based metrics.

A central goal in the cognitive sciences is the development of numerical models for mental representations of object concepts. This paper introduces Variational Interpretable Concept Embeddings (VICE), an approximate Bayesian method for embedding object concepts in a vector space using data collected from humans in a triplet odd-one-out task. VICE uses variational inference to obtain sparse, non-negative representations of object concepts with uncertainty estimates for the embedding values. These estimates are used to automatically select the dimensions that best explain the data. We derive a PAC learning bound for VICE that can be used to estimate generalization performance or determine a sufficient sample size for experimental design. VICE rivals or outperforms its predecessor, SPoSE, at predicting human behavior in the triplet odd-one-out task. Furthermore, VICE's object representations are more reproducible and consistent across random initializations, highlighting the unique advantage of using VICE for deriving interpretable embeddings from human behavior.

Collecting the large datasets needed to train deep neural networks can be very difficult, particularly for the many applications for which sharing and pooling data is complicated by practical, ethical, or legal concerns. However, it may be the case that derivative datasets or predictive models developed within individual sites can be shared and combined with fewer restrictions. Training on distributed data and combining the resulting networks is often viewed as continual learning, but these methods require networks to be trained sequentially. In this paper, we introduce distributed weight consolidation (DWC), a continual learning method to consolidate the weights of separate neural networks, each trained on an independent dataset. We evaluated DWC with a brain segmentation case study, where we consolidated dilated convolutional neural networks trained on independent structural magnetic resonance imaging (sMRI) datasets from different sites. We found that DWC led to increased performance on test sets from the different sites, while maintaining generalization performance for a very large and completely independent multi-site dataset, compared to an ensemble baseline.

Kidney segmentation is an essential step in developing any non-invasive Computer-Assisted Diagnostic (CAD) system for the early detection of acute renal rejection. This paper describes a 3-D approach for kidney segmentation from abdominal Computed Tomography (CT) images using a level set-based deformable model. Its evolution is controlled by a specially designed stochastic speed function that accounts for a shape prior and features of image intensity and spatial interactions. The shape prior is learned from the co-aligned 3-D kidney data. The current visual appearances are described with marginal gray level distributions obtained by separating their mixture over the kidney data. The spatial interactions between the kidney voxels are modeled by a 3-D 2nd-order translation and rotation variant Markov-Gibbs Random Field (MGRF) of “object-background” labels with analytically estimated potentials. The proposed approach has been evaluated on the CT data sets of 29 patients, yielding an average volumetric overlap error of 3.71%. The presented results indicate that combing CT images' characteristics into level set evolution leads to more accurate segmentation results.