Deep learning (DL) is a form of machine learning that uses hierarchical layers of abstraction to model complex structures. DL requires efficient training strategies and these are at the heart of today's successful applications which range from natural language processing to engineering and financial analysis. While deep learning has been available for several decades there were only a few practical applications until the early 2010s when the field has changed for several reasons. The renaissance is due to a number of factors, in particular 1. Hardware and software for accelerated computing (GPUs and specialized linear algebra libraries) 2. Increased size of datasets (Massive Data) 3. Efficient algorithms algorithms, such as stochastic gradient descent (SGD). The goal of our article is to provide the reader with an overview of computational aspects underlying the algorithms and hardware, which allow modern deep learning models to be implemented at scale. Many of the leading Internet companies employ DL at scale Hazelwood et al. [2017]. The most impressive accomplishment of DL is its many successful applications in research and business.

The model size of deep neural network is getting larger and larger to realize superior performance in complicated tasks. This makes it difficult to implement deep neural network in small edge-computing devices. To overcome this problem, model compression methods have been gathering much attention. However, there have been only few theoretical back-grounds that explain what kind of quantity determines the compression ability. To resolve this issue, we develop a new theoretical frame-work for model compression, and propose a new method called {\it Spectral-Pruning} based on the theory. Our theoretical analysis is based on the observation such that the eigenvalues of the covariance matrix of the output from nodes in the internal layers often shows rapid decay. We define "degree of freedom" to quantify an intrinsic dimensionality of the model by using the eigenvalue distribution and show that the compression ability is essentially controlled by this quantity. Along with this, we give a generalization error bound of the compressed model. Our proposed method is applicable to wide range of models, unlike the existing methods, e.g., ones possess complicated branches as implemented in SegNet and ResNet. Our method makes use of both "input" and "output" in each layer and is easy to implement. We apply our method to several datasets to justify our theoretical analyses and show that the proposed method achieves the state-of-the-art performance.

The current autonomous driving architecture places a heavy burden in signal processing for the graphics processing units (GPUs) in the car. This directly transforms into battery drain and lower energy efficiency, crucial factors in electric vehicles. This is due to the high bit rate of the captured video and other sensing inputs, mainly due to Light Detection and Ranging (LiDAR) sensor at the top of the car which is an essential feature in todays autonomous vehicles. LiDAR is needed to obtain a high precision map for the vehicle AI to make relevant decisions. However, this is still a quite restricted view from the car. This is the same even in the case of cars without a LiDAR such as Telsa. The existing LiDARs and the cameras have limited horizontal and vertical fields of visions. In all cases it can be argued that precision is lower, given the smaller map generated. This also results in the accumulation of a huge amount of data in the order of several TBs in a day, the storage of which becomes challenging. If we are to reduce the effort for the processing units inside the car, we need to uplink the data to edge or an appropriately placed cloud. However, the required data rates in the order of several Gbps are difficult to be met even with the advent of 5G. Therefore, we propose to have a coordinated set of LiDAR's outside at an elevation which can provide an integrated view with a much larger field of vision (FoV) to a centralized decision making body which then sends the required control actions to the vehicles with a lower bit rate in the downlink and with the required latency. The calculations we have based on industry standard equipment from several manufacturers show that this is not just a concept but a feasible system which can be implemented.

Adversarial examples are inputs to machine learning models designed to cause the model to make a mistake. They are useful for understanding the shortcomings of machine learning models, interpreting their results, and for regularisation. In NLP, however, most example generation strategies produce input text by using known, pre-specified semantic transformations, requiring significant manual effort and in-depth understanding of the problem and domain. In this paper, we investigate the problem of automatically generating adversarial examples that violate a set of given First-Order Logic constraints in Natural Language Inference (NLI). We reduce the problem of identifying such adversarial examples to a combinatorial optimisation problem, by maximising a quantity measuring the degree of violation of such constraints and by using a language model for generating linguistically-plausible examples. Furthermore, we propose a method for adversarially regularising neural NLI models for incorporating background knowledge. Our results show that, while the proposed method does not always improve results on the SNLI and MultiNLI datasets, it significantly and consistently increases the predictive accuracy on adversarially-crafted datasets -- up to a 79.6% relative improvement -- while drastically reducing the number of background knowledge violations. Furthermore, we show that adversarial examples transfer among model architectures, and that the proposed adversarial training procedure improves the robustness of NLI models to adversarial examples.

Some of Norway's largest companies are joining forces in establishing a national powerhouse for artificial intelligence. Its aim is to improve the quality and capacity for research, education and innovation in the field. Norway has a huge potential to be a pioneer in Artificial Intelligence (AI), but it needs resources and collaboration in order not to lag behind. To strengthen national efforts on artificial intelligence, Telenor, NTNU and SINTEF are inviting Norwegian businesses to partner on the new Norwegian Open AI Lab. While the Norwegian Open AI Lab will develop solutions specific to the partners' industries, it will also consider opportunities where Norway can take positions internationally.

Algorithms increasingly govern our social world, transforming data into scores or rankings that decide who gets credit, jobs, dates, policing, and much more. The field of "algorithmic accountability" has arisen to highlight the problems with such methods of classifying people, and it has great promise: Cutting-edge work in critical algorithm studies applies social theory to current events; law and policy experts seem to publish new articles daily on how artificial intelligence shapes our lives, and a growing community of researchers has developed a field known as "Fairness, Accuracy, and Transparency in Machine Learning." The social scientists, attorneys, and computer scientists promoting algorithmic accountability aspire to advance knowledge and promote justice. But what should such "accountability" more specifically consist of? At a two-day, interdisciplinary roundtable on AI ethics I recently attended, such questions featured prominently, and humanists, policy experts, and lawyers engaged in a free-wheeling discussion about topics ranging from robot arms races to computationally planned economies.

Pit bulls get a bad rap, which is especially vexing given that no one actually knows exactly what a pit bull is. There's no unified definition, because "pit bull" is not a recognized breed. But the label can have devastating consequences for dogs in shelters, who are perceived as less adoptable because of their purported heritage. In recent years, especially with the advent of genetic testing, some researchers have a new idea: just stop labelling mixed-breed dogs altogether. Researchers at Arizona State University decided to do a large-scale analysis of shelter dogs by looking at every pup that came through the doors of two animal shelters, one in Phoenix, AZ and one in San Diego, CA.

MIT CSAIL's origami robot is packaged in an ingestible ice pill. In 2013, University of Sheffield roboticist Dana Damian was doing postdoctoral research at Harvard Medical School affiliate Boston Children's Hospital when she learned of a procedure called the Foker technique. The surgery, performed on children with a rare congenital lung defect, calls for doctors to attach sutures to part of an infant's esophagus, then tie them off on the baby's back. Over time, the sutures lengthen the esophagus by pulling on it, stimulating tissue growth. Although the technique can be effective, the risk of infection and complication is high, and the baby must remain under sedation for weeks.

Speaking to the BBC this week, chief economist at the Bank of England Andy Haldane said that disruption caused by the Fourth Industrial Revolution would be "on a much greater scale" than that experienced during the First Industrial Revolution in the Victorian period. Advising that the UK required a skills revolution to counteract individuals becoming "technologically unemployed", Haldane said that training was necessary to ensure workers could leverage new job opportunities as they became available in the era of artificial intelligence. The accountancy industry is one sector primed to capitalise on the rise of artificial intelligence and machine learning. The sector has already embraced automation, with intelligent software removing the traditional compliance aspect of the accountant role. And, as the government's Making Tax Digital initiative edges ever closer, the implementation of a digital tax system has encouraged accountants across the UK to confront how they view and employ technology on a day-to-day basis.

As movies become more CGI-focused, filmmakers have to be increasingly adept at "compositing" - the process of merging foreground and background images, like placing actors on top of planes, or planets, or into fictional worlds like Black Panther's Wakanda. Making these images look realistic isn't easy. Editors have to capture the subtle aesthetic transitions between foreground and background, which can be especially difficult for intricate materials like human hair that people are used to seeing look a certain way. "The tricky thing about these images is that not every pixel solely belongs to one object," says Yagiz Aksoy, a visiting researcher at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL). "In many cases it can be hard to determine which pixels are part of the background and which are part of a specific person." Getting these details right is tedious, time-consuming and difficult for anyone but the most seasoned of editors.