Medical image registration is one of the key processing steps for biomedical image analysis such as cancer diagnosis. Recently, deep learning based supervised and unsupervised image registration methods have been extensively studied due to its excellent performance in spite of ultra-fast computational time compared to the classical approaches. In this paper, we present a novel unsupervised medical image registration method that trains deep neural network for deformable registration of 3D volumes using a cycle-consistency. Thanks to the cycle consistency, the proposed deep neural networks can take diverse pair of image data with severe deformation for accurate registration. Experimental results using multiphase liver CT images demonstrate that our method provides very precise 3D image registration within a few seconds, resulting in more accurate cancer size estimation.

LSTM based language models are an important part of modern LVCSR systems as they significantly improve performance over traditional backoff language models. Incorporating them efficiently into decoding has been notoriously difficult. In this paper we present an approach based on a combination of one-pass decoding and lattice rescoring. We perform decoding with the LSTM-LM in the first pass but recombine hypothesis that share the last two words, afterwards we rescore the resulting lattice. We run our systems on GPGPU equipped machines and are able to produce competitive results on the Hub5'00 and Librispeech evaluation corpora with a runtime better than real-time. In addition we shortly investigate the possibility to carry out the full sum over all state-sequences belonging to a given word-hypothesis during decoding without recombination.

There has been an increased interest in multimodal language processing including multimodal dialog, question answering, sentiment analysis, and speech recognition. However, naturally occurring multimodal data is often imperfect as a result of imperfect modalities, missing entries or noise corruption. To address these concerns, we present a regularization method based on tensor rank minimization. Our method is based on the observation that high-dimensional multimodal time series data often exhibit correlations across time and modalities which leads to low-rank tensor representations. However, the presence of noise or incomplete values breaks these correlations and results in tensor representations of higher rank. We design a model to learn such tensor representations and effectively regularize their rank. Experiments on multimodal language data show that our model achieves good results across various levels of imperfection.

The recent progress on capsule networks by Hinton et al. has generated considerable excitement in the machine learning community. The idea behind a capsule is inspired by a cortical minicolumn in the brain, whereby a vertically organised group of around 100 neurons receive common inputs, have common outputs, are interconnected, and may well constitute a fundamental computation unit of the cerebral cortex. However, Hinton's paper on "Matrix Capsule with EM Routing'" was unfortunately not accompanied by a release of source code, which left interested researchers attempting to implement the architecture and reproduce the benchmarks on their own. This has certainly slowed the progress of research building on this work. While writing our own implementation, we noticed several common mistakes in other open source implementations that we came across. In this paper we share some of these learnings, specifically focusing on three implementation pitfalls and how to avoid them: (1) parent capsules with only one child; (2) normalising the amount of data assigned to parent capsules; (3) parent capsules at different positions compete for child capsules. While our implementation is a considerable improvement over currently available implementations, it still falls slightly short of the performance reported by Hinton et al. (2018). The source code for this implementation is available on GitHub at the following URL: https://github.com/IBM/matrix-capsules-with-em-routing.

We will work together to foster global economic growth, while harnessing the power of technological innovation, in particular digitalization, and its application for the benefit of all. We are resolved to build a society capable of seizing opportunities, and tackling economic, social and environmental challenges, presented today and in the future, including those of demographic change. This recovery is supported by the continuation of accommodative financial conditions and stimulus measures taking effect in some countries. However, growth remains low and risks remain tilted to the downside. Most importantly, trade and geopolitical tensions have intensified. We will continue to address these risks, and stand ready to take further action. Fiscal policy should be flexible and growth-friendly while rebuilding buffers where needed and ensuring debt as a share of GDP is on a sustainable path. Monetary policy will continue to support economic activity and ensure price stability, consistent with central banks' mandates. Central bank decisions need to remain well communicated.

Neural networks are behind technologies that are revolutionizing our daily lives, such as face recognition, web searching, and medical diagnosis. These general problem solvers reach their solutions by being adapted or "trained" to capture correlations in real-world data. Having seen the success of neural networks, physicists are asking if the tools might also be useful in areas ranging from high-energy physics to quantum computing [1]. Four research groups now report on using neural network tools to tackle one of the most computationally challenging problems in condensed-matter physics--simulating the behavior of an open many-body quantum system [2–5]. This scenario describes a collection of particles--such as the qubits in a quantum computer--that both interact with each other and exchange energy with their environment.

If you've ever attended an AI conference, I bet you passed under the placid gaze of a chrome-plated humanoid, lovingly selected from an ocean of creepy robot stock images that marketing teams can't resist pasting on every billboard these days. Clearly, I'm personally guilty of using octarine-blue sci-fi art to lure weary travelers to my blog. It certainly works, which is why it's a pity that those images have next to nothing to do with AI. You'd think we'd all be more ashamed of ourselves, but don't worry, AI is too useful to go away, no matter how much we all cry wolf. Marketing folk run around trying to get your attention with sci-fi gimmicks, but the reason you'll stick around long enough to buy into AI is entirely different.

Rapid advances in computing and robotics have led to calls for government controls. Before acting, we need to distinguish among the many meanings and applications of the technology. New technologies often spur public anxiety, but the intensity of concern about the implications of advances in artificial intelligence (AI) is particularly noteworthy. Several respected scholars and technology leaders warn that AI is on the path to turning robots into a master class that will subjugate humanity, if not destroy it. Others fear that AI is enabling governments to mass produce autonomous weapons--"killing machines"--that will choose their own targets, including innocent civilians. Renowned economists point out that AI, unlike previous technologies, is destroying many more jobs than it creates, leading to major economic disruptions. There seems to be widespread agreement that AI growth is accelerating.

Interaction function (IFC), which captures interactions among items and users, is of great importance in collaborative filtering (CF). The inner product is the most popular IFC due to its success in low-rank matrix factorization. However, interactions in real-world applications can be highly complex. Many other operations (such as plus and concatenation) have also been proposed, and can possibly offer better performance than the inner product. In this paper, motivated by the success of automated machine learning, we propose to search for proper interaction functions (SIF) for CF tasks. We first design an expressive search space for SIF by reviewing and generalizing existing CF approaches. We then propose to represent the search space as a structured multi-layer perceptron, and design a stochastic gradient descent algorithm which can simultaneously update both architectures and learning parameters. Experimental results demonstrate that the proposed method can be much more efficient than popular AutoML approaches, and also obtain much better prediction performance than state-of-the-art CF approaches.

Matrix factorization methods are extensively employed to understand complex data. In this paper, we introduce the cross-product penalized component analysis (XCAN), a sparse matrix factorization based on the optimization of a loss function that allows a trade-off between variance maximization and structural preservation. The approach is based on previous developments, notably (i) the Sparse Principal Component Analysis (SPCA) framework based on the LASSO, (ii) extensions of SPCA to constrain both modes of the factorization, like co-clustering or the Penalized Matrix Decomposition (PMD), and (iii) the Group-wise Principal Component Analysis (GPCA) method. The result is a flexible modeling approach that can be used for data exploration in a large variety of problems. We demonstrate its use with applications from different disciplines.