Archives for November 21, 2018

A team of researchers from the Earthquake Research Institute, Department of Civil Engineering and Information Technology Center at the University of Tokyo, and the RIKEN Center for Computational Science and RIKEN Center for Advanced Intelligence Project in Japan were finalists for the coveted Gordon Bell Prize for outstanding achievements in high-performance computing. Tsuyoshi Ichimura together with Kohei Fujita, Takuma Yamaguchi, Kengo Nakajima, Muneo Hori and Lalith Maddegedara were praised for their simulation of earthquake physics in complex urban environments.

Earthquakes are a huge problem in many places around the world including, famously, Japan. They can be devastating and Ichimura’s team use coding prowess with the power of supercomputers to generate models for disaster mitigation and response.

Realistic earthquake simulations are difficult due to wide-ranging physical phenomena operating at different scales. This complex problem led the team to devise novel strategies involving artificial intelligence (AI) to model earthquakes in urban centers with a high degree of accuracy.

“In the field of computer science there is a big gap between AI and physics-based simulations,” said Ichimura. “We felt there was scope to enhance performance of our simulation by bridging this gap. And that feeling turned out to be true.”

Their mixed-methodology approach used AI and varying degrees of mathematical precision to create a completely new code for the simulation—with unprecedented efficiency. This new code achieved an almost fourfold increase in speed over the team’s previous incarnation.

Traditionally, physical simulations require great numerical accuracy to obtain results that correspond well with observed reality. To achieve this precision requires a lot of computing time, which consumes a great amount of power. What makes this new method unique is how the AI component of the system learns where precision is most useful and where it can be reduced without sacrificing overall accuracy, so the simulation can run in less time than if it lacked the AI.

The team’s code ran on the state-of-the-art Summit supercomputer at the Oak Ridge National Laboratory in the U.S. The researchers made this code adaptable for other uses and scalable for use on different computer systems such as the K computer at RIKEN and Piz Daint at the Swiss National Supercomputing Centre.

“Our code is an entirely new kind of problem solver, which is a frontier in this field,” concludes Ichimura. “We expect this new code will find its way into a new generation of physical simulators. We hope this helps people better understand, predict and prepare for earthquakes.”

Classic winner and millionaire 3-year-old Cloud Computing has been retired from racing and will enter stud in 2019 at B. Wayne Hughes’ Spendthrift Farm near Lexington, the farm announced Nov. 20.

The dark bay colt will become the newest stallion to participate in Spendthrift’s “Share The Upside” program.

“Cloud Computing is one of those horses that sells himself immediately when you see him,” said Ned Toffey, Spendthrift general manager, of the Preakness Stakes (G1) winner. “He’s a classic winner, but he’s also the picture of what a classic horse is supposed to look like. For him to go on and win the Preakness over Classic Empire less than 100 days after making his debut, that’s pretty special, and it speaks to his quality. There’s a lot to like about Cloud Computing, and we believe breeders are going to love what they see.”

The 4-year-old son of Maclean’s Music is out of a grade 2-placed A.P. Indy mare Quick Temper. His second dam, Halo America, won the 1997 Apple Blossom Handicap (G1) and three other graded stakes.

The Share The Upside fee for Cloud Computing will be $8,500 for one year. Additionally, breeders must breed a mare in 2020 on a complimentary basis. After the breeder has a live foal in 2020, pays the stud fee, and breeds a mare back, he or she will earn a lifetime breeding right beginning in 2021. For breeders not interested in the Share The Upside program, Cloud Computing will be offered for $7,500 on a standard stands and nurses contract. As always, Share The Upside opportunities are limited and offered on a first-come, first-served basis.

Campaigned by Klaravich Stables and William Lawrence, Cloud Computing jumped straight onto the Kentucky Derby Trail with a runner-up finish in the Gotham Stakes (G3) three weeks after breaking his maiden. In his next start, he overcame a poor break and wide trip to rally for third in the Wood Memorial Stakes Presented by NYRA Bets (G2).

Despite qualifying for the Kentucky Derby Presented by Yum! Brands (G1), trainer Chad Brown instead pointed Cloud Computing to the Preakness because he had run three times in eight weeks. The colt rewarded his connections with their first classic victory when he ran down 2-year-old champion Classic Empire in the stretch and stopped the clock in 1:55.98. The 102 Beyer Speed Figure he earned for this performance—which matches the winning Beyer figures for recent Preakness winners American Pharoah and Lookin At Lucky —exceeds the Beyers earned by Justify and Exaggerator in the middle jewel of the Triple Crown.

The classic win—which was over a field that also included Kentucky Derby winner Always Dreaming and multimillionaire Gunnevera, among others—came only 98 days after Cloud Computing’s career debut, making him one of racing’s fastest millionaires in recent history.

“Cloud Computing is one of the best-looking horses I’ve ever had walk into my shed row,” said Brown. “We always had tremendous confidence in this horse, and he showed why in the Preakness. To give up so much seasoning to a champion like Classic Empire, he showed real brilliance to beat that colt. He’s my first classic winner, and he’s very special to us.”

Cloud Computing retires with earnings of $1,125,200. He was purchased by Mike Ryan as agent for $200,000 out of the Hill ‘n’ Dale Sales Agency consignment to the Keeneland September Yearling Sale, making him the highest-priced yearling from the first crop of Maclean’s Music.

“When I saw him at Hill ‘n’ Dale in August of 2015 right before the (Keeneland September) sale, I was awestruck. It was instant,” said Ryan. “In fact, I thought he was one of the best yearlings I had seen in 2015. He’s LeBron James. He’s got it all: size, strength, substance, quality, and tons of class.

“He’s such an outstanding physical specimen that I’d be very confident he will reproduce himself. You’ll go a long way before you find a better-looking horse than Cloud Computing. I think when people see him, they will be tremendously impressed with his physique and presence.”

Snap is reportedly set to release a new version of Spectacles with an aluminum design that packs in two cameras. Through the Snapchat app, you may be able to add augmented reality overlay effects in videos you capture with the updated Spectacles. The $350 frames will cost more than double the first version, which arrived in 2016, and will be on sale by the end of the year, according to Cheddar.

Following a $40 million write-down on the first version after Snap overestimated demand (it ordered around 800,000 pairs), it has been more conservative with orders of the device. For the second version of Spectacles, which were released earlier this year, Snap reportedly ordered 35,000 pairs, and 52,000 of a tweaked set it released in September. This time around, it’s hedging its bets further by ordering around 24,000, according to the report.

Snap is set to release its latest Spectacles after watching its stock price plummet 50 percent this year, and in the wake of a damaging report about how CEO Evan Speigel was “dismissive” to board members and spends much of his time during board meetings using Snapchat on his phone. The latest report indicates that Snap is willing to lose money on the Spectacles side of the business for now, with the hope of breaking even in 2020. Snap, meanwhile, is facing an investigation over allegations it misled investors over user metrics.

In 2017, over 1 crore students appeared for class 12 exams. These students are expected to grow up to be successful doctors, engineers, lawyers, or chartered accountants. In fact, right from school, they have faced relentless pressure to shine in their exams and secure a seat in ivy league colleges. However, a very small number of these students have had access to a quality education, knowledgeable and well-trained teachers, modern infrastructure, and up to date reference material.

Over the last decade, technology has seen unprecedented growth. Apart from social media apps, better mobile phone cameras, faster laptops, technology has solved problems that seemed unsolvable just a few years ago. One of these problems is providing access to high quality, personalised education to every student in India.

Over the last 5 years, edtech apps have been harnessing the power of Artificial Intelligence, Machine Learning, Big Data, and Natural Language Processing to completely revolutionise education. Here is how they do it:

1. Personalised learning and feedback:

We believe that every child is unique, and has a unique set of learning needs. If these needs are addressed, the child’s learning process sees a sharp, positive growth. However, a classroom teacher cannot teach a class of 60 in 60 different ways.

Advanced technology can be used to adjust the pace of learning content to suit the learning speed of individual students. Due to this, students will neither get bored by slow teaching nor will they play catch up with a fast-paced lecture.

Moreover, instead of simply awarding a grade, it can also be used to provide in-depth, insightful feedback to students at an individual level. It can provide a subject wise strength and weakness analysis, along with highlighting topics where they have been going wrong more often than others. It can also show classmates where they stand amongst their peers.

With enough time, it is capable of providing qualitative feedback for more complex answering formats like essays, long answers, and opinion articles. This will provide a big boost to their overall learning curve.

2. Adaptive learning paths:

Advanced technology can observe the learning pattern through the app and test the behaviour of every student. Through this data, it can gauge concepts that a student finds difficult or easy. It can use this data and various insights to restructure the learning direction based on the abilities of a student. It can create unique learning paths customised for every child.

This technology will also be capable of creating personalised improvement approaches to help students plan for competitive exams. It can personalise learning across any and every age group, IQ, and interest. Due to this, children will always learn in a format created for them, instead of trying to fit their learning technique into a predefined box.

3. 24*7 support to students:

Natural language processing (NLP) combined with optical character recognition (OCR) among other algorithms can solve any and every doubt raised by a student instantly, round the clock.

All a student has to do is click a photo of the doubt and send to a chat on the doubts app or helpline. The bot will then scan through millions of data points to either give an exact solution and explanation to the student, or that of a similar question.

The advantage of having such a bot is that it is steadily developing. It will understand and collect more information with every doubt and query uploaded by its students. This reduces the dependency of the availability of expert tutors. Advanced technology can provide reliable and precise answers in just a few seconds at any point in time.

Technology is set to dominate the way education is addressed. With its team of engineers and entrepreneurs, India can become a leading powerhouse in this field. It can use technology to ensure that within a very short span of time, high-quality education is available for any and every child in the country.