Earlier than Sep-12-2017


U.S. to Unveil Streamlined Autonomous Vehicle Guidelines Tuesday

U.S. News

The House measure would allow automakers to field up to 25,000 vehicles without meeting existing auto safety standards in the first year. Over three years, the cap would rise to 100,000 vehicles annually. Automakers would be required to provide regulators with safety assessments of their systems, but would not have to get federal approval to put autonomous cars on the road.


Controversial AI system can 'tell if you're gay'

Daily Mail

Digitally scanning contours of the face, cheekbones, nose and chin the computer made hosts of measurements of the ratios between the different facial features. Digitally scanning contours of the face, cheekbones, nose and chin the computer made hosts of measurements of the ratios between the different facial features. In a leading article, the Economist warns: 'In countries where homosexuality is a crime, software which promises to infer sexuality from a face is an alarming prospect.' In a leading article, the Economist warns: 'In countries where homosexuality is a crime, software which promises to infer sexuality from a face is an alarming prospect.'


GM might beat its competitors to driverless car production

Engadget

While a large number of companies are working on self-driving cars, including Google/Waymo, Uber, Audi and Tesla, it might be General Motors that beats them all. In a blog post, GM's CEO of Cruise Automation, Kyle Vogt, says that his team has created the "world's first mass-producible car designed to operate without a driver." They also meet the safety requirements the company believes are necessary to work without a driver and will be part of the San Francisco fleet in the next few weeks. GM's Cruise team wants to see a fully autonomous car that won't require a "safety driver" like current autonomous cars do.


How Machine Learning and AI could eventually replace development work

#artificialintelligence

ML and AI assistants will perform automated code reviews the moment we push code to repositories. Those bot assistants will become more intelligent and they will perform automated code reviews and comment the code in case it makes no sense based on the training data from similar code or codebases. We will declare our functional and non functional requirements and our technical requirements and the ML assistant will train models to produce code that abides to those constraints. Applications will be able to deploy themselves automatically, write test cases in case of an error automatically and push the test cases to the repositories.


Should The Use of Machine Learning in Healthcare Be Embraced or Met With Skepticism?

#artificialintelligence

Natural language processing (NLP) of unstructured free-text content, such as that found in electronic health record (EHR) clinical notes, used to diagnose disease. Using data provided by the patient upon admission to the hospital and prior to the collection of vital signs, laboratory results or patient history to classify that patient's risk for a particular disease state. Natural language processing (NLP) of unstructured free-text content, such as that found in electronic health record (EHR) clinical notes, used to diagnose disease. Using data provided by the patient upon admission to the hospital and prior to the collection of vital signs, laboratory results or patient history to classify that patient's risk for a particular disease state.


Mind-reading AI is the newest defense against cyber attacks

#artificialintelligence

Criminals are holding data and information hostage using tools that require almost no skill, and once they have your files it's too late to react. This "mind reading" is accomplished initially by data gathering – we read the data generated by a variety of tools inside the organization – which is then enriched by Internet data sources which yield more signals and cues. No one signal lets us read the attackers' mind, but we connected the dots to generate intention. Empow's patented mind-reading AI won't tell you what you're thinking, but if it brings bigger guns in the fight against cyber-attacks it'll be far more important than a crystal ball.


Getting Started with Watson Virtual Agent

#artificialintelligence

IBM Watson, Twilio & Node.JS: Making a Twilio Chatbot powered by Watson Conversation-Icecream Sundae - Duration: 32:55. IBM Watson 7,167 views Watson Natural Language Understanding Service Overview - Duration: 3:06. Chris & Chris on Chatbots 27,925 views Watson Machine Learning: Get Started With IBM Watson Machine Learning in IBM Bluemix - Duration: 1:23. Watson Machine Learning: Get Started With IBM Watson Machine Learning in IBM Bluemix - Duration: 1:23.


Hackers Have Already Started to Weaponize Artificial Intelligence

#artificialintelligence

But make no mistake--modern tools like machine intelligence and neural networks are a form of artificial intelligence, and to believe otherwise is something we do at our own peril; if we dismiss or ignore the power of these tools, we may be blindsided by those who are eager to exploit AI's full potential, hackers included. While it can be argued that automation is fundamentally unintelligent (conversely, a case can be made that some forms of automation, particularly those involving large sets of complex tasks, are indeed a form of intelligence), it's the prospect of a machine intelligence orchestrating these automated tasks that's particularly alarming. An AI can produce complex and highly targeted scripts at a rate and level of sophistication far beyond any individual human hacker. In addition to the criminal activities already described, AIs could be used to target vulnerable populations, perform rapid-fire hacks, develop intelligent malware, and so on.


What the Hell is "Tensor" in "TensorFlow"? – Hacker Noon

@machinelearnbot

It is also represented with a Rank, like in Matrix. If you think like me, that it is an upgrade of vectors like vectors is an upgrade to scalars. Both the quantities Scalar and Vector are Tensors. The number of dimensions used to represent the data is known as its Rank.


Kaggle instacart (top2%) feature engineering and solution overview - Jacques Peeters's blog

@machinelearnbot

This blog post aims at showing what kind of feature engineering can be achieved in order to improve machine learning models. In this blog post i'll detail my general approach (in a machine learning way) and the feature engineering work which was done. Feature engineering is the oil allowing machine learning models to shine. In my opinion feature engineering and data wrangling is more important than models!