This paper begins with considering the identification of sparse linear time-invariant networks described by multivariable ARX models. Such models possess relatively simple structure thus used as a benchmark to promote further research. With identifiability of the network guaranteed, this paper presents an identification method that infers both the Boolean structure of the network and the internal dynamics between nodes. Identification is performed directly from data without any prior knowledge of the system, including its order. The proposed method solves the identification problem using Maximum a posteriori estimation (MAP) but with inseparable penalties for complexity, both in terms of element (order of nonzero connections) and group sparsity (network topology). Such an approach is widely applied in Compressive Sensing (CS) and known as Sparse Bayesian Learning (SBL). We then propose a novel scheme that combines sparse Bayesian and group sparse Bayesian to efficiently solve the problem. The resulted algorithm has a similar form of the standard Sparse Group Lasso (SGL) while with known noise variance, it simplifies to exact re-weighted SGL. The method and the developed toolbox can be applied to infer networks from a wide range of fields, including systems biology applications such as signaling and genetic regulatory networks.

As deep neural networks continue to revolutionize various application domains, there is increasing interest in making these powerful models more understandable and interpretable, and narrowing down the causes of good and bad predictions. We focus on recurrent neural networks (RNNs), state of the art models in speech recognition and translation. Our approach to increasing interpretability is by combining an RNN with a hidden Markov model (HMM), a simpler and more transparent model. We explore various combinations of RNNs and HMMs: an HMM trained on LSTM states; a hybrid model where an HMM is trained first, then a small LSTM is given HMM state distributions and trained to fill in gaps in the HMM's performance; and a jointly trained hybrid model. We find that the LSTM and HMM learn complementary information about the features in the text.

While state-of-the-art kernels for graphs with discrete labels scale well to graphs with thousands of nodes, the few existing kernels for graphs with continuous attributes, unfortunately, do not scale well. To overcome this limitation, we present hash graph kernels, a general framework to derive kernels for graphs with continuous attributes from discrete ones. The idea is to iteratively turn continuous attributes into discrete labels using randomized hash functions. We illustrate hash graph kernels for the Weisfeiler-Lehman subtree kernel and for the shortest-path kernel. The resulting novel graph kernels are shown to be, both, able to handle graphs with continuous attributes and scalable to large graphs and data sets. This is supported by our theoretical analysis and demonstrated by an extensive experimental evaluation.

Computed tomography (CT) generates a stack of cross-sectional images covering a region of the body. The visual assessment of these images for the identification of potential abnormalities is a challenging and time consuming task due to the large amount of information that needs to be processed. In this article we propose a deep artificial neural network architecture, ReCTnet, for the fully-automated detection of pulmonary nodules in CT scans. The architecture learns to distinguish nodules and normal structures at the pixel level and generates three-dimensional probability maps highlighting areas that are likely to harbour the objects of interest. Convolutional and recurrent layers are combined to learn expressive image representations exploiting the spatial dependencies across axial slices. We demonstrate that leveraging intra-slice dependencies substantially increases the sensitivity to detect pulmonary nodules without inflating the false positive rate. On the publicly available LIDC/IDRI dataset consisting of 1,018 annotated CT scans, ReCTnet reaches a detection sensitivity of 90.5% with an average of 4.5 false positives per scan. Comparisons with a competing multi-channel convolutional neural network for multi-slice segmentation and other published methodologies using the same dataset provide evidence that ReCTnet offers significant performance gains.

Causal inference concerns the identification of cause-effect relationships between variables. However, often only linear combinations of variables constitute meaningful causal variables. For example, recovering the signal of a cortical source from electroencephalography requires a well-tuned combination of signals recorded at multiple electrodes. We recently introduced the MERLiN (Mixture Effect Recovery in Linear Networks) algorithm that is able to recover, from an observed linear mixture, a causal variable that is a linear effect of another given variable. Here we relax the assumption of this cause-effect relationship being linear and present an extended algorithm that can pick up non-linear cause-effect relationships. Thus, the main contribution is an algorithm (and ready to use code) that has broader applicability and allows for a richer model class. Furthermore, a comparative analysis indicates that the assumption of linear cause-effect relationships is not restrictive in analysing electroencephalographic data.

We propose Turing Learning, a novel system identification method for inferring the behavior of natural or artificial systems. Turing Learning simultaneously optimizes two populations of computer programs, one representing models of the behavior of the system under investigation, and the other representing classifiers. By observing the behavior of the system as well as the behaviors produced by the models, two sets of data samples are obtained. The classifiers are rewarded for discriminating between these two sets, that is, for correctly categorizing data samples as either genuine or counterfeit. Conversely, the models are rewarded for 'tricking' the classifiers into categorizing their data samples as genuine. Unlike other methods for system identification, Turing Learning does not require predefined metrics to quantify the difference between the system and its models. We present two case studies with swarms of simulated robots and prove that the underlying behaviors cannot be inferred by a metric-based system identification method. By contrast, Turing Learning infers the behaviors with high accuracy. It also produces a useful by-product - the classifiers - that can be used to detect abnormal behavior in the swarm. Moreover, we show that Turing Learning also successfully infers the behavior of physical robot swarms. The results show that collective behaviors can be directly inferred from motion trajectories of individuals in the swarm, which may have significant implications for the study of animal collectives. Furthermore, Turing Learning could prove useful whenever a behavior is not easily characterizable using metrics, making it suitable for a wide range of applications.

International Business Machines Corp. plans to purchase consultancy Promontory Financial Group LLC., the two companies said Thursday, creating a new subsidiary dubbed'Watson Financial Services.' The companies didn't disclose financial details of the deal, which they said is subject to regulatory approvals. The idea, they said, is to combine Promontory's financial regulatory expertise with Watson, IBM IBM 0.61 % 's artificial intelligence computer system, to help banks meet ever-rising regulatory expectations in areas such as anti-money-laundering detection systems, consumer complaint databases, and so-called stress tests. Financial regulatory requirements are "rapidly outstripping the capacity of humans to keep up," a joint press release from the companies said. Privately held Promontory was founded in 2001 by Eugene Ludwig, a former U.S. Comptroller of the Currency, and has hired a small army of former regulators and government officials to advise financial firms.

During practices, American football coaches typically stay on the sidelines, grim-faced, as they order their players through drills. But during an afternoon this past May, in the cavernous training facility for the Pittsburgh Steelers, head coach Mike Tomlin couldn't resist getting in on the action. As a human-size robot sped over the artificial turf, the grinning coach ran onto the field and tackled it. The MVP, or Mobile Virtual Player, was designed to take precisely this kind of hit--the sort of jarring blow that, inflicted repeatedly, can injure the brains of human players. American football has been rocked by controversy over the last decade, as it has become clear that the repeated collisions inherent to the sport are giving players concussions and sometimes causing debilitating and permanent brain trauma. In response, the U.S. National Football League (NFL) has altered rules and contributed millions to medical research. Meanwhile, the same head-injury concerns have found even greater resonance in college and youth football.

The Terminator isn't arriving anytime soon, but concern is growing that artificial intelligence is already so pervasive in society--and getting more so all the time--that there needs to be more focus on how it's being used and potentially misused (even if by accident). Aside from futuristic killer robots, there are already real dangers ranging from faulty autonomous cars to algorithms used in hiring or recruiting that have an inadvertent bias against women or ethnic groups. The giants of artificial intelligence, especially as it affects consumers and businesses, have just joined together to form a nonprofit called the Partnership on AI, with founding members Amazon, DeepMind/Google, Facebook, IBM, and Microsoft. It's the latest effort to keep a collective eye on how AI is developed and used. OpenAI, founded in December 2015, has a similar goal of conducing research and conferences to promote responsible use of AI.

REDMOND, Washington -- Sept. 29, 2016 -- Microsoft Corp. announced on Thursday it has formed the Microsoft AI and Research Group, bringing together Microsoft's world-class research organization with more than 5,000 computer scientists and engineers focused on the company's AI product efforts. The new group will be led by computer vision luminary Harry Shum, a 20-year Microsoft veteran whose career has spanned leadership roles across Microsoft Research and Bing engineering. Microsoft is dedicated to democratizing AI for every person and organization, making it more accessible and valuable to everyone and ultimately enabling new ways to solve some of society's toughest challenges. Today's announcement builds on the company's deep focus on AI and will accelerate the delivery of new capabilities to customers across agents, apps, services and infrastructure. In addition to Shum's existing leadership team, several of the company's engineering leaders and teams will join the newly formed group including Information Platform, Cortana and Bing, and Ambient Computing and Robotics teams led by David Ku, Derrick Connell and Vijay Mital, respectively.