Donald Trump has hammered Amazon.com Inc.'s share price in recent weeks by trying to pick fights online with Chief Executive Officer Jeff Bezos. But the actual object of his Twitter onslaught is the domain of Jeff Wilke, Bezos' right-hand man for most of the past 18 years. Wilke runs Amazon's worldwide consumer division, meaning he's in charge of both selling people stuff and figuring out how to deliver the items as efficiently as possible. Often, the company uses the U.S. Postal Service.

When it comes to robots stealing our jobs, there's one task that plenty of people would be happy to relinquish: assembling flat-pack furniture. The day we can handball that job off is a step closer. In Science Robotics today, a trio of roboticists from Nanyang Technological University in Singapore unveiled a semi-autonomous pair of robotic arms that can put together the frame of an IKEA chair. Made from off-the-shelf gear, their invention can look at chair parts scattered around, grasp and lift the right bit, carefully insert wooden pins in pre-drilled holes and slot the pieces together. And while the robot's repertoire of furniture-building skills is limited -- it can't yet screw in metal screws, for instance -- the technology could soon be ready for jobs that require human-like dexterity, such as electronics and aircraft manufacturing.

Spotify is beginning to roll out a re-design of its mobile app to select users of its free-tier service. The changes, which include more control over playlists, will allow Spotify to make its free version behave more like a Premium account in a bid to boost user numbers. The update is currently in testing and has only been rolled out to a small amount of users. The search page has had a slight overhaul, with colourful tabs indicating different genres and a glimpse at what artists are contained in each playlist. The revamped bottom bar is a new'Premium' button to allow users to easily upgrade to the premium version of the app Rumours of the mobile update were unveiled last week by Bloomberg.

How does Artificial Neural Network (ANN) algorithm work?

In the largest survey of its kind so far, researchers have looked at the "DNA" of over 340,000 stars in our home galaxy, to identify the stars with the same chemical composition as the sun. Stars with the same chemistry originated in the same star clusters, and the survey would help astronomers find siblings of the sun before the Milky Way tore apart the stellar nursery where they all formed. The research is led by astronomers from Australia, supported by colleagues from Europe, and is called the Galactic Archaeology survey, or GALAH. It "is a large-scale stellar spectroscopic survey of the Milky Way and designed to deliver chemical information" about a large number of stars. The announcement Tuesday about 342,682 stars was the first major public release of data by GALAH, which will eventually look at over one million stars.

Which statement do you believe? Robots will wipe out our jobs. AI and robotics will make everything free. These extreme viewpoints are both vying for our attention. Singularity University, which aims to solve our global grand challenges through exponential technologies, widely reports that AI is the world's cure.

The Bharat Bandh on April 2 called by Dalit groups was a result of the successful use of artificial intelligence (AI), the geographic information system (GIS) and big data, according to a core group of US-based Dalit activists. The protest, which surprised many because of its sudden eruption, was organised against the alleged dilution of the SC/ST (Prevention of Atrocities) Act. New Jersey-based Deelip Mhaske, who heads a secretive network of some 100 overseas Dalit activists based in the US, the UK, Australia and the Middle East said the protest shows that Dalits no longer need to be affiliated to any political establishment. "A new era for Dalit politics has begun by adopting artificial intelligence, the GIS and social media," Mhaske told PTI in an interview as he gave an insight into how Dalit groups were mobilised during the protest. Most of these Dalit activists in the secretive group are from the IT sector and work in some of the top info-tech companies.

A good number of us shout at our laptops when they misbehave, often to no avail. Perhaps soon they will listen. Could we one day teach them--much like we do children or pets--how to behave? For the majority of human history, we have survived and flourished based on our ability to learn. Today's machines learn--Siri perks up at the sound of your voice, traffic lights react to the flux of cars--but only in limited ways.

Very recently proximal policy optimization (PPO) algorithms have been proposed as first-order optimization methods for effective reinforcement learning. While PPO is inspired by the same learning theory that justifies trust region policy optimization (TRPO), PPO substantially simplifies algorithm design and improves data efficiency by performing multiple epochs of \emph{clipped policy optimization} from sampled data. Although clipping in PPO stands for an important new mechanism for efficient and reliable policy update, it may fail to adaptively improve learning performance in accordance with the importance of each sampled state. To address this issue, a new surrogate learning objective featuring an adaptive clipping mechanism is proposed in this paper, enabling us to develop a new algorithm, known as PPO-$\lambda$. PPO-$\lambda$ optimizes policies repeatedly based on a theoretical target for adaptive policy improvement. Meanwhile, destructively large policy update can be effectively prevented through both clipping and adaptive control of a hyperparameter $\lambda$ in PPO-$\lambda$, ensuring high learning reliability. PPO-$\lambda$ enjoys the same simple and efficient design as PPO. Empirically on several Atari game playing tasks and benchmark control tasks, PPO-$\lambda$ also achieved clearly better performance than PPO.

Traditional programs take data as input and produces data as output. However, a machine learning algorithm takes data as input but produces a program as an output. This machine generated program can now take new data, process it and produce output data. In simple words, it gives computers the capability to extract knowledge from data and store it for future judgment. Machine learning is the subfield of computer science that, according to Arthur Samuel, gives "computers the ability to learn without being explicitly programmed".