Breast cancer is one of the most common cancers in women worldwide, and early detection can significantly reduce the mortality rate of breast cancer. It is crucial to take multi-scale information of tissue structure into account in the detection of breast cancer. And thus, it is the key to design an accurate computer-aided detection (CAD) system to capture multi-scale contextual features in a cancerous tissue. In this work, we present a novel graph convolutional neural network for histopathological image classification of breast cancer. The new method, named multi-scale graph wavelet neural network (MS-GWNN), leverages the localization property of spectral graph wavelet to perform multi-scale analysis. By aggregating features at different scales, MS-GWNN can encode the multi-scale contextual interactions in the whole pathological slide. Experimental results on two public datasets demonstrate the superiority of the proposed method. Moreover, through ablation studies, we find that multi-scale analysis has a significant impact on the accuracy of cancer diagnosis.

Traditional approaches to computer security isolate systems from the outside world through a combination of firewalls, passwords, data encryption, and other access control measures. In contrast, dataset creators often invite the outside world in -- data-hungry neural network models are built by harvesting information from anonymous and unverified sources on the web. Such open-world dataset creation methods can be exploited in several ways. Outsiders can passively manipulate datasets by placing corrupted data on the web and waiting for data harvesting bots to collect them. Active dataset manipulation occurs when outsiders have the privilege of sending corrupted samples directly to a dataset aggregator such as a chatbot, spam filter, or database of user profiles. Adversaries may also inject data into systems that rely on federated learning, in which models are trained on a diffuse network of edge devices that communicate periodically with a central server. In this case, users have complete control over the training data and labels seen by their device, in addition to the content of updates sent to the central server.

Anomaly detection aims to find instances that are considered unusual and is a fundamental problem of data science. Recently, deep anomaly detection methods were shown to achieve superior results particularly in complex data such as images. Our work focuses on deep one-class classification for anomaly detection which learns a mapping only from the normal samples. However, the non-linear transformation performed by deep learning can potentially find patterns associated with social bias. The challenge with adding fairness to deep anomaly detection is to ensure both making fair and correct anomaly predictions simultaneously. In this paper, we propose a new architecture for the fair anomaly detection approach (Deep Fair SVDD) and train it using an adversarial network to de-correlate the relationships between the sensitive attributes and the learned representations. This differs from how fairness is typically added namely as a regularizer or a constraint. Further, we propose two effective fairness measures and empirically demonstrate that existing deep anomaly detection methods are unfair. We show that our proposed approach can remove the unfairness largely with minimal loss on the anomaly detection performance. Lastly, we conduct an in-depth analysis to show the strength and limitations of our proposed model, including parameter analysis, feature visualization, and run-time analysis.

Many concepts have been proposed for meta learning with neural networks (NNs), e.g., NNs that learn to control fast weights, hyper networks, learned learning rules, and meta recurrent neural networks (Meta RNNs). Our Variable Shared Meta Learning (VS-ML) unifies the above and demonstrates that simple weight-sharing and sparsity in an NN is sufficient to express powerful learning algorithms. A simple implementation of VS-ML called Variable Shared Meta RNN allows for implementing the backpropagation learning algorithm solely by running an RNN in forward-mode. It can even meta-learn new learning algorithms that improve upon backpropagation, generalizing to different datasets without explicit gradient calculation.

Despite the latest prevailing success of deep neural networks (DNNs), several concerns have been raised against their usage, including the lack of intepretability the gap between DNNs and other well-established machine learning models, and the growingly expensive computational costs. A number of recent works [1], [2], [3] explored the alternative to sequentially stacking decision tree/random forest building blocks in a purely feed-forward way, with no need of back propagation. Since decision trees enjoy inherent reasoning transparency, such deep forest models can also facilitate the understanding of the internaldecision making process. This paper further extends the deep forest idea in several important aspects. Firstly, we employ a probabilistic tree whose nodes make probabilistic routing decisions, a.k.a., soft routing, rather than hard binary decisions.Besides enhancing the flexibility, it also enables non-greedy optimization for each tree. Second, we propose an innovative topology learning strategy: every node in the ree now maintains a new learnable hyperparameter indicating the probability that it will be a leaf node. In that way, the tree will jointly optimize both its parameters and the tree topology during training. Experiments on the MNIST dataset demonstrate that our empowered deep forests can achieve better or comparable performance than [1],[3] , with dramatically reduced model complexity. For example,our model with only 1 layer of 15 trees can perform comparably with the model in [3] with 2 layers of 2000 trees each.

We introduce twin neural network (TNN) regression. This method predicts differences between the target values of two different data points rather than the targets themselves. The solution of a traditional regression problem is then obtained by averaging over an ensemble of all predicted differences between the targets of an unseen data point and all training data points. Whereas ensembles are normally costly to produce, TNN regression intrinsically creates an ensemble of predictions of twice the size of the training set while only training a single neural network. Since ensembles have been shown to be more accurate than single models this property naturally transfers to TNN regression. We show that TNNs are able to compete or yield more accurate predictions for different data sets, compared to other state-of-the-art methods. Furthermore, TNN regression is constrained by self-consistency conditions. We find that the violation of these conditions provides an estimate for the prediction uncertainty.

Solar flares are caused by magnetic eruptions in active regions (ARs) on the surface of the sun. These events can have significant impacts on human activity, many of which can be mitigated with enough advance warning from good forecasts. To date, machine learning-based flare-prediction methods have employed physics-based attributes of the AR images as features; more recently, there has been some work that uses features deduced automatically by deep learning methods (such as convolutional neural networks). We describe a suite of novel shape-based features extracted from magnetogram images of the Sun using the tools of computational topology and computational geometry. We evaluate these features in the context of a multi-layer perceptron (MLP) neural network and compare their performance against the traditional physics-based attributes. We show that these abstract shape-based features outperform the features chosen by the human experts, and that a combination of the two feature sets improves the forecasting capability even further.

Deep neural networks are the default choice of learning models for computer vision tasks. Extensive work has been carried out in recent years on explaining deep models for vision tasks such as classification. However, recent work has shown that it is possible for these models to produce substantially different attribution maps even when two very similar images are given to the network, raising serious questions about trustworthiness. To address this issue, we propose a robust attribution training strategy to improve attributional robustness of deep neural networks. Our method carefully analyzes the requirements for attributional robustness and introduces two new regularizers that preserve a model's attribution map during attacks. Our method surpasses state-of-the-art attributional robustness methods by a margin of approximately 3% to 9% in terms of attribution robustness measures on several datasets including MNIST, FMNIST, Flower and GTSRB.

This work presents a single-step deep-learning framework for longitudinal image analysis, coined Segis-Net. To optimally exploit information available in longitudinal data, this method concurrently learns a multi-class segmentation and nonlinear registration. Segmentation and registration are modeled using a convolutional neural network and optimized simultaneously for their mutual benefit. An objective function that optimizes spatial correspondence for the segmented structures across time-points is proposed. We applied Segis-Net to the analysis of white matter tracts from N=8045 longitudinal brain MRI datasets of 3249 elderly individuals. Segis-Net approach showed a significant increase in registration accuracy, spatio-temporal segmentation consistency, and reproducibility comparing with two multistage pipelines. This also led to a significant reduction in the sample-size that would be required to achieve the same statistical power in analyzing tract-specific measures. Thus, we expect that Segis-Net can serve as a new reliable tool to support longitudinal imaging studies to investigate macro- and microstructural brain changes over time.

Stacked convolution and pooling layers enable Convolutional Neural Networks (CNNs) to learn hierarchical representation of grid-like data[1], where the convolution extracts local patterns of the data and the pooling layers reduce the computation cost by compressing the data shape. Because both of the two operations are defined on planar grids in Euclidean domains, they cannot be directly employed in graph data, which is a more general case and widely used in fields of chemical molecules, drug design and social networks. Learning the hierarchical representation of graph is a challenging problem and one of the solutions is to extend the convolution and pooling to graph. Graph convolution includes spatial and spectral methods[2, 3], both of which can be seen as a message passing process on multi-hop graphs. For implementation on graphs of massive number of nodes, 1-order convolution, represented by GCN and GAT[4, 5], become increasingly popular, but abandon part of ability to capturing complex graph pattern.