Several recent works have shown that state-of-the-art classifiers are vulnerable to worst-case (i.e., adversarial) perturbations of the datapoints. On the other hand, it has been empirically observed that these same classifiers are relatively robust to random noise. In this paper, we propose to study a \textit{semi-random} noise regime that generalizes both the random and worst-case noise regimes. We propose the first quantitative analysis of the robustness of nonlinear classifiers in this general noise regime. We establish precise theoretical bounds on the robustness of classifiers in this general regime, which depend on the curvature of the classifier's decision boundary. Our bounds confirm and quantify the empirical observations that classifiers satisfying curvature constraints are robust to random noise. Moreover, we quantify the robustness of classifiers in terms of the subspace dimension in the semi-random noise regime, and show that our bounds remarkably interpolate between the worst-case and random noise regimes. We perform experiments and show that the derived bounds provide very accurate estimates when applied to various state-of-the-art deep neural networks and datasets. This result suggests bounds on the curvature of the classifiers' decision boundaries that we support experimentally, and more generally offers important insights onto the geometry of high dimensional classification problems.

Here's How Google Will Use A.I. to Help Fight Cancer How UC Berkeley's New Center Could Prevent a Military A.I. Apocalypse Beauty.AI App the 1st international beauty contest judged by AI A treasure hunter went missing in the Rocky Mountains, and a computer algorithm found him ... Drive.ai wants to give self-driving cars more brainpower, personality

Google's DeepMind's Health initiative will soon work to explore how artificial intelligence could save hours and hours of precious time in the treatment of oral, head and neck cancers, the company announced on Tuesday. Whatever it is Zuckerberg is calling his personal home AI system. Yesterday he updated an audience in Rome about his personal project this year. He had Facebook (FB) engineers work on a computerized domestic assistant as his personal challenge. The AI controls his AC, lighting, opens his security gate, and can even make him toast.

When I first became interested in using deep learning for computer vision I found it hard to get started. There were only a couple of open source projects available, they had little documentation, were very experimental, and relied on a lot of tricky-to-install dependencies. A lot of new projects have appeared since, but they're still aimed at vision researchers, so you'll still hit a lot of the same obstacles if you're approaching them from outside the field. In this article -- and the accompanying webcast -- I'm going to show you how to run a pre-built network, and then take you through the steps of training your own. I've listed the steps I followed to set up everything toward the end of the article, but because the process is so involved, I recommend you download a Vagrant virtual machine that I've pre-loaded with everything you need.

Google DeepMind is launching a project to reduce the time it takes doctors to prepare treatment for head and neck cancers. Alphabet's London-based artificial intelligence division has partnered with the UK's National Health Service and will be conducting the research in coordination with the University College London Hospital. Head and neck cancers are hard to plan treatment for because of their close proximity to important parts of the body. Before any kind of radiation treatment, clinicians will prepare a detailed map of where radiation will be administered on a patient in order to avoid damaging surrounding tissue. DeepMind says planning can take doctors up to four hours for head and neck cancers, and it hopes that by applying machine learning it will be able to automate parts of the process and reduce that planning time down to an hour.

Nvidia has been a leader in producing the technology behind high-quality graphics for years, but the company is now betting on a different future. With the rapid advances in self-driving vehicles, warehouse robots, diagnostic assistants, and speech and facial recognition, there's plenty of reasons for companies to be excited about deep-learning-based artificial intelligence (AI). Nvidia is one such technology firm that is extremely bullish in this space. In a recent conversation with CXOtoday, Vishal Dhupar, Managing Director-South Asia, Nvidia said that the company is moving ahead with its AI-focused hardware, software, and solutions for the enterprise. As such data scientists in both industry and academia have been using graphics processing units (GPUs) for machine learning to make groundbreaking innovations across a variety of applications including image classification, video analytics, speech recognition and natural language processing or NLP. In particular, Deep Learning, the use of sophisticated, multi-level "deep" neural networks to create systems that can perform feature detection from massive amounts of unlabeled training data, is an area where Nvidia is significantly investing in recent quarters.

The OpenAI site is centered around an Open Source project and community involving artificial intelligence. The term "Open Source" means that the source code for the project is available for free and can be used by others free of charge. Artificial Intelligence refers to the general aim of intelligent computing, making machines think and learn. The project itself is the creation of a set of tools that are considered to be models of human intelligence. These tools are intended to be integrated into programs or used stand alone for research.

Google DeepMindhas announced it is working on a project to improve treatment on head and neck cancers, its third major collaboration with the NHS. The London-based AI research arm of the online search firm is partnering with University College London Hospital in an attempt to improve the scans available for radiotherapists by using machine learning. The project will use anonymised scans from up to 700 former patients. Radiotherapy works by bombarding cancerous cells with radiation to kill them, while minimising damage to the healthy cells around them. Clinicians target the treatment through a process called "segmentation": literally drawing around different parts of the patient's anatomy on scans, letting the radiotherapy machines know which tissue to target and which tissue to leave.

Drive.ai uses deep learning to teach self-driving cars – and to give them a voice Drive.ai wants to help autonomous cars talk with the people around them Drive.ai AI-based startup Banihal helps find a perfect match with neuroscience-based recommendation ...

AI machines are racist & it's all our fault - report