Artificial Intelligence (AI) keeps on exciting the world with its many uses and potentials. Through deep learning (also called deep neural networks), we now have computers and robots which are capable of thinking independently like humans. And this quality makes them even more efficient than us! Image recognition, natural language processing, and machine translation may not be perfect yet. But, when it comes to data analysis, we know that computers are way, way faster and more precise than we are.

The artificial intelligence arms race continues, as the largest tech companies explore new ways accelerate AI workloads for cloud platforms. The appetite for more computing horsepower is following several tracks, with major investment in graphics processors (GPUs) as well as custom ASIC chips. Microsoft has been a leader in using FPGAs (Field Programmable Gate Arrays) to accelerate its cloud and AI workloads. This week Microsoft unveiled Project Brainwave, a deep learning acceleration platform based on its collaboration with Intel on FPGA computing. Microsoft says Project Brainwave represents a "major leap forward" in cloud-based deep learning performance, and intends to bring the technology to its Windows Azure cloud computing platform.

Compressed sensing (CS) is a sampling theory that allows reconstruction of sparse (or compressible) signals from an incomplete number of measurements, using of a sensing mechanism implemented by an appropriate projection matrix. The CS theory is based on random Gaussian projection matrices, which satisfy recovery guarantees with high probability; however, sparse ternary {0, -1, +1} projections are more suitable for hardware implementation. In this paper, we present a deep learning approach to obtain very sparse ternary projections for compressed sensing. Our deep learning architecture jointly learns a pair of a projection matrix and a reconstruction operator in an end-to-end fashion. The experimental results on real images demonstrate the effectiveness of the proposed approach compared to state-of-the-art methods, with significant advantage in terms of complexity.

With the availability of large databases and recent improvements in deep learning methodology, the performance of AI systems is reaching or even exceeding the human level on an increasing number of complex tasks. Impressive examples of this development can be found in domains such as image classification, sentiment analysis, speech understanding or strategic game playing. However, because of their nested non-linear structure, these highly successful machine learning and artificial intelligence models are usually applied in a black box manner, i.e., no information is provided about what exactly makes them arrive at their predictions. Since this lack of transparency can be a major drawback, e.g., in medical applications, the development of methods for visualizing, explaining and interpreting deep learning models has recently attracted increasing attention. This paper summarizes recent developments in this field and makes a plea for more interpretability in artificial intelligence. Furthermore, it presents two approaches to explaining predictions of deep learning models, one method which computes the sensitivity of the prediction with respect to changes in the input and one approach which meaningfully decomposes the decision in terms of the input variables. These methods are evaluated on three classification tasks.

Image denoising techniques are essential to reducing noise levels and enhancing diagnosis reliability in low-dose computed tomography (CT). Machine learning based denoising methods have shown great potential in removing the complex and spatial-variant noises in CT images. However, some residue artifacts would appear in the denoised image due to complexity of noises. A cascaded training network was proposed in this work, where the trained CNN was applied on the training dataset to initiate new trainings and remove artifacts induced by denoising. A cascades of convolutional neural networks (CNN) were built iteratively to achieve better performance with simple CNN structures. Experiments were carried out on 2016 Low-dose CT Grand Challenge datasets to evaluate the method's performance.

Humans take advantage of real world symmetries for various tasks, yet capturing their superb symmetry perception mechanism with a computational model remains elusive. Motivated by a new study demonstrating the extremely high inter-person accuracy of human perceived symmetries in the wild, we have constructed the first deep-learning neural network for reflection and rotation symmetry detection (Sym-NET), trained on photos from MS-COCO (Microsoft-Common Object in COntext) dataset with nearly 11K consistent symmetry-labels from more than 400 human observers. We employ novel methods to convert discrete human labels into symmetry heatmaps, capture symmetry densely in an image and quantitatively evaluate Sym-NET against multiple existing computer vision algorithms. On CVPR 2013 symmetry competition testsets and unseen MS-COCO photos, Sym-NET significantly outperforms all other competitors. Beyond mathematically well-defined symmetries on a plane, Sym-NET demonstrates abilities to identify viewpoint-varied 3D symmetries, partially occluded symmetrical objects, and symmetries at a semantic level.

Assessment of the impact of natural disasters on infrastructure systems is of importance toward four main objectives: (1) Planning for actions that eliminate or reduce the long-term risk to human life and infrastructure systems (e.g.[2]); (2) Disaster preparation or adjustment, which aims to reduce the risk of damages and injuries while enabling the capability to cope with the temporary disruption of the infrastructure systems (e.g.[3]); (3) Development of effective emergency response strategies (e.g.[4]); and (4) Post-disaster recovery planning (e.g.[5]). These four are, respectively, known as the mitigation, preparedness, response, and recovery practices. A variety of analytical [6], simulation [7-11], and optimization [12] approaches are proposed in the literature for hazard reliability analysis of infrastructure systems. A comprehensive literature review on transportation infrastructure system performance in disasters is provided in [13]. Simulation-based reliability assessment of large infrastructure systems are often computationally intractable or expensive due to the large number of network components, complex network topology, statistical dependence between component failures, and uncertainties in the hazard models. This will impose limitations on design optimization or sensitivity analysis of these systems. Alternatively, a more efficient response assessment for large infrastructure systems can be made possible by using approximate surrogates [14]. Surrogates are fast models that approximately describe the relationship between the system inputs and outputs and serve as a substitute for more expensive simulation tools. If the response evaluated by the reference expensive model is denoted by f (x), a surrgate seeks to provide a global approximate function f (x).

Then this paragraph is self-explanatory. Deep Learning methods don't work easily with reinforcement learning like they do in supervised/unsupervised learning. Most DL applications have involved huge training datasets with accurate samples and labels. Or in unsupervised learning, the target cost function is still quite quite convenient to work with. But in RL, there's a catch -- as you know, RL involves rewards which could be delayed many time steps into the future (for example it takes several moves to knock the opponent's queen in chess, and each of those moves doesn't return the same immediate reward as the final move, EVEN IF one of those moves might be more important than the final move). The rewards could also be noisy -- for instance, sometimes the points for a particular move are slightly random and not easily predictable!

Summary: Reinforcement Learning (RL) is likely to be the next big push in artificial intelligence. But the concept of modeling in RL is very different from our statistical techniques and deep learning. In this two part series we'll take a look at the basics of RL models, how they're built and used. In the next part, we'll address some of the complexities that make development a challenge. Now that we have pretty much conquered speech, text, and image processing with deep neural nets, it's time to turn our attention to what comes next. It's likely that the next most important area of development for AI will be reinforcement learning (RL).

Walt Disney Co. (NYSE:DIS) presented at a conference last month its newly developed artificial-intelligence (AI) technology for tracking the facial reactions of theatergoers as they watch a movie. The tech could help the entertainment giant make even better movies and could also have other applications across its empire. Disney tapped an NVIDIA (NASDAQ:NVDA) graphics processing unit (GPU) to play a starring role in the development of its new tech. Here's what you should know. Disney's new factorized variational autoencoders (FVAE) tech falls within the branch of AI called deep learning, which aims to mimic human thought processes. In deep learning, an artificial neural network is trained how to think or make inferences, and then it's deployed where it makes inferences from new data, which could be images, speech, and so on.