Optimus Prime--the software engine, not the Autobot overlord--was born in a basement under a West Elm furniture store on University Avenue in Palo Alto. Starting two years ago, a band of artificial-intelligence acolytes within Salesforce escaped the towering headquarters with the goal of crazily multiplying the impact of the machine learning models that increasingly shape our digital world--by automating the creation of those models. As shoppers checked out sofas above their heads, they built a system to do just that. Scott Rosenberg is an editor at Backchannel. Sign up to get Backchannel's weekly newsletter. They named it after the Transformers leader because, as one participant recalls, "machine learning is all about transforming data."

While most of us think we are consciously creating all of our thoughts and decisions through rational processes, research in the field of neuroscience has actually told us something else might be going on. Making decisions or forming opinions like this may not actually be the reality when it comes to our reactions to new situations, ideas or even products – really, in any situation in which you react emotionally. According to neuroscience, that's because most emotional decisions of this kind are made unconsciously. When you see something new, and you feel a certain way about it, do you think about why you should feel the way you do first – or do you just feel a certain way about that thing or situation and then try to find reasons for the reaction you are having? Most likely, you thought about your feelings a bit, and decided why you felt the way you did.

Aida is the perfect employee: always courteous, always learning and, as she says, "always at work, 24/7, 365 days a year." Aida, of course, is not a person but a virtual customer-service representative that SEB AB, one of Sweden's biggest banks, is rolling out. The goal is to give the actual humans more time to engage in more complex tasks. After blazing a trail in online and digital banking, Sweden's financial industry is now emerging as a pioneer in the use of artificial intelligence (AI). Besides Aida at SEB, there's Nova, which is a chatbot Nordea Bank is introducing at its life and pensions unit in Norway.

Computer Vision generates mathematical models from images; Computer Graphics draws in images from models and lastly image processing takes image as an input and gives an image at the output. Computer Vision is an overlapping field drawing on concepts from areas such as artificial intelligence, digital image processing, machine learning, deep learning, pattern recognition, probabilistic graphical models, scientific computing and a lot of mathematics. Watch these videos and alongside implementing the learned concepts and algorithms by following GaTech Prof. James Hays' projects of his Computer Vision class. Have a quick go through Building Machine Learning Systems with Python and Python Machine Learning.

Many modern data analysis environments allow for code-free creation of advanced analytics workflows. The advantages are obvious: more casual users, who cannot possibly stay on top of the complexity of working in a programming environment, are empowered to use existing workflows as templates and modify them to fit their needs, thus creating complex analytics protocols that they would never have been able to create in a programming environment. At the same time, these visual environments serve as an excellent means for documentation purposes. Instead of having to read code, the visual representation intuitively explains which steps have been performed and – in most environments at least ‒ the configuration of each module is self-explanatory as well. This enables a broad set of intuitively reusable workflows to be built up capturing the data scientists' wisdom.

I attended the first and largest of its kind, UN ITU XPRIZE Artificial Intelligence (AI) for Good Global Summit held 7-9 June 2017 in Geneva touching 300 plus million, influencing 1 billion, and meaningfully impacting 7.2 billion via more than 30 global media, social media, and television crews, including a documentary crew. If you have not done so already you want to get over to the summit website and view the rich collection of content there created by the top leaders and minds in AI, civil society, 21 UN organizations, governments, industry, academia, and media. Just a few of the notable speakers and participants were: Peter Norvig head of research at Google, Peter Lee Corporate Vice President of Microsoft AI and Research, Margaret Chan Director General of the World Health Organization (WHO), Pedro Domingos Professor of Computer Science and Engineering at the University of Washington, Jü rgen Schmidhuber, Scientific Director, Swiss AI Lab, IDSIA; Professor of AI, USI & SUPSI, Switzerland; President of NNAISENSE, Andy Chen Board Chair at IEEE Computer Society, and Vicki Hanson, President of ACM, and Distinguished Professor of Computing at the Rochester Institute of Technology. The Summit sought to speed the development of, and access to, AI solutions to address global challenges ranging from poverty, hunger, health and education to equality and environmental sustainability. In my last article, I challenged everyone including all in the business world to deliberately embrace AI in a way that results in good outcomes.

Researchers have imaged whole live insects using a technique called X-ray micro-computed tomography (micro-CT). Imaging whole live insects is challenging because of the radiation doses required to obtain the images and the movement of the insects, compromising image quality - so most research involving involving invertebrates imaging requires animal sacrifice. But in a new study, researchers used carbon dioxide gas to anesthetize the insects to prevent them from moving during 3D visualization, and according to the researchers, the method had very little impact on the insects' longevity. In order to image live insects, the researchers say it's crucial to fully immobilize them, deliver a low-enough radiation dose to allow for repeated scans of the same individuals for time-relevant studies, and provide adequate image resolution and quality to distinguish internal structures. The researchers, based at the University of Western Ontario, scanned Colorado potato beetles (Leptinotarsa decemlineata) and true armyworms (Pseudaletia unipuncta).

Dogs have a toddler's level of emotional awareness and understanding, new research suggests. The study shows that dogs display a level of emotional intelligence comparable with that of a human two to three-year-old. Evidence from recent brain scan studies strengthens the case for banning scientific tests carried out on man's best friend, the researchers claim. Dogs have a toddler's level of emotional awareness and understanding, new research suggests. The team conducted a review of research involving functional magnetic resonance imaging (MRI) brain scans of dogs.

Semantic segmentation was seen as a challenging computer vision problem few years ago. Due to recent advancements in deep learning, relatively accurate solutions are now possible for its use in automated driving. In this paper, the semantic segmentation problem is explored from the perspective of automated driving. Most of the current semantic segmentation algorithms are designed for generic images and do not incorporate prior structure and end goal for automated driving. First, the paper begins with a generic taxonomic survey of semantic segmentation algorithms and then discusses how it fits in the context of automated driving. Second, the particular challenges of deploying it into a safety system which needs high level of accuracy and robustness are listed. Third, different alternatives instead of using an independent semantic segmentation module are explored. Finally, an empirical evaluation of various semantic segmentation architectures was performed on CamVid dataset in terms of accuracy and speed. This paper is a preliminary shorter version of a more detailed survey which is work in progress.

Convolutional Neural Networks have been a subject of great importance over the past decade and great strides have been made in their utility for producing state of the art performance in many computer vision problems. However, the behavior of deep networks is yet to be fully understood and is still an active area of research. In this work, we present an intriguing behavior: pre-trained CNNs can be made to improve their predictions by structurally perturbing the input. We observe that these perturbations - referred as Guided Perturbations - enable a trained network to improve its prediction performance without any learning or change in network weights. We perform various ablative experiments to understand how these perturbations affect the local context and feature representations. Furthermore, we demonstrate that this idea can improve performance of several existing approaches on semantic segmentation and scene labeling tasks on the PASCAL VOC dataset and supervised classification tasks on MNIST and CIFAR10 datasets.