Which authors of this paper are endorsers? Disable MathJax (What is MathJax?)

Brain tumor is considered as one of the deadliest and most common form of cancer both in children and in adults. Consequently, determining the correct type of brain tumor in early stages is of significant importance to devise a precise treatment plan and predict patient's response to the adopted treatment. In this regard, there has been a recent surge of interest in designing Convolutional Neural Networks (CNNs) for the problem of brain tumor type classification. However, CNNs typically require large amount of training data and can not properly handle input transformations. Capsule networks (referred to as CapsNets) are brand new machine learning architectures proposed very recently to overcome these shortcomings of CNNs, and posed to revolutionize deep learning solutions. Of particular interest to this work is that Capsule networks are robust to rotation and affine transformation, and require far less training data, which is the case for processing medical image datasets including brain Magnetic Resonance Imaging (MRI) images.

"It is possible for ASICs over time to be successful in the deep-learning world," Mosesmann said. "However, we are of the opinion that at this stage in a multidecade product cycle it is just too early to'fix' the hardware, given that there is a plethora of deep-learning frameworks (Tensorflow, Caffee, MXNet, …

I often have to write in different languages and was shocked to find that overnight Google Translate went from being a joke to "knowing" even my best foreign language better than me. The reason: an AI program called "Deep Learning."

Neural-Network Quantum States have been recently introduced as an Ansatz for describing the wave function of quantum many-body systems. We show that there are strong connections between Neural-Network Quantum States in the form of Restricted Boltzmann Machines and some classes of Tensor-Network states in arbitrary dimensions. In particular we demonstrate that short-range Restricted Boltzmann Machines are Entangled Plaquette States, while fully connected Restricted Boltzmann Machines are String-Bond States with a nonlocal geometry and low bond dimension. These results shed light on the underlying architecture of Restricted Boltzmann Machines and their efficiency at representing many-body quantum states. String-Bond States also provide a generic way of enhancing the power of Neural-Network Quantum States and a natural generalization to systems with larger local Hilbert space. We compare the advantages and drawbacks of these different classes of states and present a method to combine them together. This allows us to benefit from both the entanglement structure of Tensor Networks and the efficiency of Neural-Network Quantum States into a single Ansatz capable of targeting the wave function of strongly correlated systems. While it remains a challenge to describe states with chiral topological order using traditional Tensor Networks, we show that Neural-Network Quantum States and their String-Bond States extension can describe a lattice Fractional Quantum Hall state exactly. In addition, we provide numerical evidence that Neural-Network Quantum States can approximate a chiral spin liquid with better accuracy than Entangled Plaquette States and local String-Bond States. Our results demonstrate the efficiency of neural networks to describe complex quantum wave functions and pave the way towards the use of String-Bond States as a tool in more traditional machine-learning applications.

We develop three efficient approaches for generating visual explanations from 3D convolutional neural networks (3D-CNNs) for Alzheimer's disease classification. One approach conducts sensitivity analysis on hierarchical 3D image segmentation, and the other two visualize network activations on a spatial map. Visual checks and a quantitative localization benchmark indicate that all approaches identify important brain parts for Alzheimer's disease diagnosis. Comparative analysis show that the sensitivity analysis based approach has difficulty handling loosely distributed cerebral cortex, and approaches based on visualization of activations are constrained by the resolution of the convolutional layer. The complementarity of these methods improves the understanding of 3D-CNNs in Alzheimer's disease classification from different perspectives.

The performance of automatic speech recognition (ASR) systems can be significantly compromised by previously unseen conditions, which is typically due to a mismatch between training and testing distributions. In this paper, we address robustness by studying domain invariant features, such that domain information becomes transparent to ASR systems, resolving the mismatch problem. Specifically, we investigate a recent model, called the Factorized Hierarchical Variational Autoencoder (FHVAE). FHVAEs learn to factorize sequence-level and segment-level attributes into different latent variables without supervision. We argue that the set of latent variables that contain segment-level information is our desired domain invariant feature for ASR. Experiments are conducted on Aurora-4 and CHiME-4, which demonstrate 41% and 27% absolute word error rate reductions respectively on mismatched domains.

We use deep learning to model interactions across two or more sets of objects, such as user-movie ratings or protein-drug bindings. The canonical representation of such interactions is a matrix (or tensor) with an exchangeability property: the encoding's meaning is not changed by permuting rows or columns. We argue that models should hence be Permutation Equivariant (PE): constrained to make the same predictions across such permutations. We present a parameter-sharing scheme and prove that it could not be made any more expressive without violating PE. This scheme yields three benefits. First, we demonstrate performance competitive with the state of the art on multiple matrix completion benchmarks. Second, our models require a number of parameters independent of the numbers of objects, and thus scale well to large datasets. Third, models can be queried about new objects that were not available at training time, but for which interactions have since been observed. We observed surprisingly good generalization performance on this matrix extrapolation task, both within domains (e.g., new users and new movies drawn from the same distribution used for training) and even across domains (e.g., predicting music ratings after training on movie ratings).

Tremendous advances in natural language processing (NLP) have been enabled by novel deep neural network architectures and word embeddings. Historically, convolutional neural network (CNN)[1, 2] and recurrent neural network (RNN)[3, 4] topologies have competed to provide state-of-the-art results for NLP tasks, ranging from text classification to reading comprehension. CNNs identify and aggregate patterns with increasing feature sizes, reflecting our common practice of identifying patterns, literal or idiomatic, for understanding language; they are thus adept at tasks involving key phrase identification. RNNs instead construct a representation of sentences by successively updating their understanding of the sentence as they read new words, appealing to the formally sequential and rule-based construction of language. While both networks display great efficacy at certain tasks [5], RNNs tend to be the more versatile, have emerged as the clear victor in, e.g., language translation [6, 7, 8], and are typically more capable of identifying important contextual points through attention mechanisms for, e.g., reading comprehension [9, 10, 11, 12].

Building effective neural networks requires many design choices. These include the network topology, optimization procedure, regularization, stability methods, and choice of pre-trained parameters. This design is time consuming and requires expert input. Automatic Machine Learning aims automate this process using hyperparameter optimization. However, automatic model building frameworks optimize performance on each task independently, whereas human experts leverage prior knowledge when designing a new network. We propose Transfer Automatic Machine Learning, a method to accelerate network design using knowledge of prior tasks. For this, we build upon reinforcement learning architecture design methods to support parallel training on multiple tasks and transfer the search strategy to new tasks. Tested on NLP and Image classification tasks, Transfer Automatic Machine Learning reduces convergence time over single-task methods by almost an order of magnitude on 13 out of 14 tasks. It achieves better test set accuracy on 10 out of 13 tasks NLP tasks and improves performance on CIFAR-10 image recognition from 95.3% to 97.1%.