The open letter argues that “once developed, lethal autonomous weapons will permit armed conflict to be fought at a scale greater than ever, and at timescales faster than humans can comprehend…we do not have long to act. Once this Pandora’s box is opened, it will be hard to close.”
The FT (paywall) discusses how Goldman Sachs has expanded its algorithmic corporate bond trading programme, more than trebling the number of securities it quotes since last summer to more than 7,000 — and is now eyeing an expansion into areas such as junk bonds later this year.
The bank’s algorithm scrapes publicly-available pricing data for thousands of bonds to automatically generate firm, tradable prices for investors. Earlier this year it broke into the ranks of the top-three dealers on MarketAxess in US investment grade odd-lots — defined as smaller slivers of debt below $1m, according to Goldman Sachs.
That science fiction future where robots can do what people and animals do may be closer than you think. Marc Raibert, founder of Boston Dynamics, is developing advanced robots that can gallop like a cheetah, negotiate 10 inches of snow, walk upright on two legs and even open doors and deliver packages. Join Raibert for a live demo of SpotMini, a nimble robot that maps the space around it, handles objects, climbs stairs — and could soon be helping you out around the house.
To try and make better sense of the issue of hate crimes, Google and ProPublica have teamed up to create an artificial-intelligence powered tool that parses through news articles, in order to build a bigger picture of where hate crimes are occurring across the country. Here’s how it works:
The Documenting Hate News Index — built by the Google News Lab, data visualisation studio Pitch Interactive and ProPublica — takes a raw feed of Google News articles from the past six months and uses the Google Cloud Natural Language API to create a visual tool to help reporters find the news happening all over the country, from Oklahoma to Florida, California to Kentucky. It’s a constantly-updating snapshot of data from this year, one which is valuable as a starting point to reporting on this area of news.
British scientists have developed the world’s smallest surgical robot which could transform everyday operations for tens of thousands of patients. From a converted pig shed in the Cambridgeshire countryside, a team of 100 scientists and engineers have used low-cost technology originally developed for mobile phones and space industries to create the first robotic arm specifically designed to carry out keyhole surgery.
The robot, called Versius, mimics the human arm and can be used to carry out a wide range of laparoscopic procedures – including hernia repairs, colorectal operations, and prostate and ear, nose and throat surgery – in which a series of small incisions are made to circumvent the need for traditional open surgery. This reduces complications and pain after surgery and speeds up recovery times for patients.
Even though Chinese AI companies are relatively strong in specific applications of AI, AI chips manufacturing has so far been the stronghold of U.S. companies. Only around 7.55% of AI investments in China were invested in processor and chip developers. That number is around 31% in the U.S., according to Tencent Research Institution.
Cambricon was founded in 2016 by Chen Tianshi, a professor at the Institute of Computing Technology at the Chinese Academy Of Sciences. Last year, the company launched a processor, named Cambricon-1A, which it claims to be the world’s first commercial chip for deep learning. The product can be applied in a number of fields, such as smartphones, security, drones, wearable devices, and autonomous driving.
Using demographic filtering, Brendan O’Connor, an assistant professor at the University of Massachusetts, Amherst, and one of his graduate students, Su Lin Blodgett, collected 59.2 million tweets with a high probability of containing African-American slang or vernacular. They then tested several natural-language processing tools on this data set to see how they would treat these statements. They found that one popular tool classified these posts as Danish with a high level of confidence.
The UMass researchers presented their work at a workshop dedicated to exploring the issue of bias in AI. The event, Fairness and Transparency in Machine Learning, was part of a larger data-science conference this year, but it will become a stand-alone conference itself in 2018.
Andrey Nikishaev has put together a how-to guide, walking you through all the aspects of machine learning from simple linear regressions to the latest neural networks, and you will learn not only how to use them but also how to build them from scratch. The guide assumes some basic level of understanding of Python.
The guide includes courses, tutorials/projects,
and FAQs to get you going.
It’s standard practice to use watermarks on the assumption that they prevent consumers from accessing the clean images, ensuring there will be no unauthorized or unlicensed use. However, in “On The Effectiveness Of Visible Watermarks” recently presented at the 2017 Computer Vision and Pattern Recognition Conference (CVPR 2017), Google researchers show that a computer algorithm can get past this protection and remove watermarks automatically, giving users unobstructed access to the clean images the watermarks are intended to protect.
As often done with vulnerabilities discovered in operating systems, applications or protocols, they want to disclose this vulnerability and propose solutions in order to help the photography and stock image communities adapt and better protect its copyrighted
content and creations.