Neural Tangents is a library designed to enable research into infinite-width neural networks. It provides a high-level API for specifying complex and hierarchical neural network architectures. These networks can then be trained and evaluated either at finite-width as usual or in their infinite-width limit. Infinite-width networks can be trained analytically using exact Bayesian inference or using gradient descent via the Neural Tangent Kernel. Additionally, Neural Tangents provides tools to study gradient descent training dynamics of wide but finite networks in either function space or weight space. The entire library runs out-of-the-box on CPU, GPU, or TPU. All computations can be automatically distributed over multiple accelerators with near-linear scaling in the number of devices. Neural Tangents is available at www.github.com/google/neural-tangents. We also provide an accompanying interactive Colab notebook.

Modern deep neural networks can achieve high accuracy when the training distribution and test distribution are identically distributed, but this assumption is frequently violated in practice. When the train and test distributions are mismatched, accuracy can plummet. Currently there are few techniques that improve robustness to unforeseen data shifts encountered during deployment. In this work, we propose a technique to improve the robustness and uncertainty estimates of image classifiers. We propose AugMix, a data processing technique that is simple to implement, adds limited computational overhead, and helps models withstand unforeseen corruptions. AugMix significantly improves robustness and uncertainty measures on challenging image classification benchmarks, closing the gap between previous methods and the best possible performance in some cases by more than half.

Normalizing flows provide a general mechanism for defining expressive probability distributions, only requiring the specification of a (usually simple) base distribution and a series of bijective transformations. There has been much recent work on normalizing flows, ranging from improving their expressive power to expanding their application. We believe the field has now matured and is in need of a unified perspective. In this review, we attempt to provide such a perspective by describing flows through the lens of probabilistic modeling and inference. We place special emphasis on the fundamental principles of flow design, and discuss foundational topics such as expressive power and computational trade-offs. We also broaden the conceptual framing of flows by relating them to more general probability transformations. Lastly, we summarize the use of flows for tasks such as generative modeling, approximate inference, and supervised learning.

Deep ensembles have been empirically shown to be a promising approach for improving accuracy, uncertainty and out-of-distribution robustness of deep learning models. While deep ensembles were theoretically motivated by the bootstrap, non-bootstrap ensembles trained with just random initialization also perform well in practice, which suggests that there could be other explanations for why deep ensembles work well. Bayesian neural networks, which learn distributions over the parameters of the network, are theoretically well-motivated by Bayesian principles, but do not perform as well as deep ensembles in practice, particularly under dataset shift. One possible explanation for this gap between theory and practice is that popular scalable approximate Bayesian methods tend to focus on a single mode, whereas deep ensembles tend to explore diverse modes in function space. We investigate this hypothesis by building on recent work on understanding the loss landscape of neural networks and adding our own exploration to measure the similarity of functions in the space of predictions. Our results show that random initializations explore entirely different modes, while functions along an optimization trajectory or sampled from the subspace thereof cluster within a single mode predictions-wise, while often deviating significantly in the weight space. We demonstrate that while low-loss connectors between modes exist, they are not connected in the space of predictions. Developing the concept of the diversity--accuracy plane, we show that the decorrelation power of random initializations is unmatched by popular subspace sampling methods.

Machine learning methods have recently achieved high-performance in biomedical text analysis. However, a major bottleneck in the widespread application of these methods is obtaining the required large amounts of annotated training data, which is resource intensive and time consuming. Recent progress in self-supervised learning has shown promise in leveraging large text corpora without explicit annotations. In this work, we built a self-supervised contextual language representation model using BERT, a deep bidirectional transformer architecture, to identify radiology reports requiring prompt communication to the referring physicians. We pre-trained the BERT model on a large unlabeled corpus of radiology reports and used the resulting contextual representations in a final text classifier for communication urgency. Our model achieved a precision of 97.0%, recall of 93.3%, and F-measure of 95.1% on an independent test set in identifying radiology reports for prompt communication, and significantly outperformed the previous state-of-the-art model based on word2vec representations.

Machine learning has started to be deployed in fields such as healthcare and finance, which propelled the need for and growth of privacy-preserving machine learning (PPML). We propose an actively secure four-party protocol (4PC), and a framework for PPML, showcasing its applications on four of the most widely-known machine learning algorithms -- Linear Regression, Logistic Regression, Neural Networks, and Convolutional Neural Networks. Our 4PC protocol tolerating at most one malicious corruption is practically efficient as compared to the existing works. We use the protocol to build an efficient mixed-world framework (Trident) to switch between the Arithmetic, Boolean, and Garbled worlds. Our framework operates in the offline-online paradigm over rings and is instantiated in an outsourced setting for machine learning. Also, we propose conversions especially relevant to privacy-preserving machine learning. The highlights of our framework include using a minimal number of expensive circuits overall as compared to ABY3. This can be seen in our technique for truncation, which does not affect the online cost of multiplication and removes the need for any circuits in the offline phase. Our B2A conversion has an improvement of $\mathbf{7} \times$ in rounds and $\mathbf{18} \times$ in the communication complexity. In addition to these, all of the special conversions for machine learning, e.g. Secure Comparison, achieve constant round complexity. The practicality of our framework is argued through improvements in the benchmarking of the aforementioned algorithms when compared with ABY3. All the protocols are implemented over a 64-bit ring in both LAN and WAN settings. Our improvements go up to $\mathbf{187} \times$ for the training phase and $\mathbf{158} \times$ for the prediction phase when observed over LAN and WAN.

Image synthesis is currently one of the most addressed image processing topic in computer vision and deep learning fields of study. Researchers have tackled this problem focusing their efforts on its several challenging problems, e.g. image quality and size, domain and pose changing, architecture of the networks, and so on. Above all, producing images belonging to different domains by using a single architecture is a very relevant goal for image generation. In fact, a single multi-domain network would allow greater flexibility and robustness in the image synthesis task than other approaches. This paper proposes a novel architecture and a training algorithm, which are able to produce multi-domain outputs using a single network. A small portion of a dataset is intentionally used, and there are no hard-coded labels (or classes). This is achieved by combining a conditional Generative Adversarial Network (cGAN) for image generation and a Meta-Learning algorithm for domain switch, and we called our approach MetalGAN. The approach has proved to be appropriate for solving the multi-domain problem and it is validated on facial attribute transfer, using CelebA dataset.

Recent work on deep neural network pruning has shown there exist sparse subnetworks that achieve equal or improved accuracy, training time, and loss using fewer network parameters when compared to their dense counterparts. Orthogonal to pruning literature, deep neural networks are known to be susceptible to adversarial examples, which may pose risks in security- or safety-critical applications. Intuition suggests that there is an inherent trade-off between sparsity and robustness such that these characteristics could not co-exist. We perform an extensive empirical evaluation and analysis testing the Lottery Ticket Hypothesis with adversarial training and show this approach enables us to find sparse, robust neural networks. Code for reproducing experiments is available here: https://github.com/justincosentino/robust-sparse-networks.

Computer-Aided Drug Discovery research has proven to be a promising direction in drug discovery. In recent years, Deep Learning approaches have been applied to problems in the domain such as Drug-Target Interaction Prediction and have shown improvements over traditional screening methods. An existing challenge is how to represent compound-target pairs in deep learning models. While several representation methods exist, such descriptor schemes tend to complement one another in many instances, as reported in the literature. In this study, we propose a multi-view architecture trained adversarially to leverage this complementary behavior by integrating both differentiable and predefined molecular descriptors. We conduct experiments on clinically relevant benchmark datasets to demonstrate the potential of our approach.

This work is a reproducibility study of the paper of Antoniou and Storkey [2019], published at NeurIPS 2019. Our results are in parts similar to the ones reported in the original paper, supporting the central claim of the paper that the proposed novel method, called Self-Critique and Adapt (SCA), improves the performance of MAML++. The conducted additional experiments on the Caltech-UCSD Birds 200 dataset confirm the superiority of SCA compared to MAML++. In addition, the reproduced paper suggests a novel high-end version of MAML++ for which we could not reproduce the same results. We hypothesize that this is due to the many implementation details that were omitted in the original paper.