NYU Data Science newsletter – January 26, 2016

NYU Data Science Newsletter features journalism, research papers, events, tools/software, and jobs for January 26, 2016

GROUP CURATION: N/A

 
Data Science News



How the artificial intelligence revolution was born in a Vancouver hotel | Financial Post

Financial Post


from January 22, 2016

Mel Silverman walked over to a whiteboard and picked up a marker, listing all the academic disciplines that the band of renegade scientists asking him for money represented.

Assembled there 12 years ago at Vancouver’s Metropolitan Hotel was a group of about 15 people, ranging from computer scientists to biologists to experimental engineers. What united them was their interest in a concept that was, at the time, generally perceived as the domain of the lunatic fringe.

They believed it was possible to teach a machine to learn the same way a child does, through artificial neural networks that mimic the function of the human brain.

 

Working in Tandem at Tandon

NYU Tandon School of Engineering


from January 19, 2016

It’s not uncommon for writers and artists to seek residencies—programs that provide physical facilities, networking opportunities, and conditions conducive to creativity. In exchange, the institutions offering those residencies may get the right to exhibit any completed artwork, a ready source of guest lecturers, and the prestige of being affiliated with acknowledged talents in their fields.

Now in its third year, the Faculty Engineer in Residence (F-EIR) program at the NYU School of Engineering allows for a similar sort of symbiosis. Professors from various disciplines, many of them with extensive entrepreneurial experience of their own, make themselves available to advise start-up companies within the university’s burgeoning incubator system; in return, the incubator companies provide Tandon students with fellowships, mentoring, and the chance to apply classroom learning to real-world problems.

 

What a Million Syllabuses Can Teach Us – The New York Times

The New York Times, SundayReview, Cristobal Young


from January 22, 2016

Over the past two years, we and our partners at the Open Syllabus Project (based at the American Assembly at Columbia) have collected more than a million syllabuses from university websites. We have also begun to extract some of their key components — their metadata — starting with their dates, their schools, their fields of study and the texts that they assign.

This past week, we made available online a beta version of our Syllabus Explorer, which allows this database to be searched. Our hope and expectation is that this tool will enable people to learn new things about teaching, publishing and intellectual history.

 

NYU Wireless To Build Test Bed for 5G

Campus Technology


from January 20, 2016

NYU Wireless, New York University’s multidisciplinary academic research center for wireless networking theories and techniques, is building an advanced programmable platform to support the development of millimeter wave (mmWave) wireless communication, also known as 5G.

5G is the next generation of cellular networking technology and has the potential to support data connection speeds exceeding 10 gigabits per second (Gbps), a thousand times faster than current 4G data rates. The mmWave spectrum could also “provide 200 times the capacity of all of today’s cellular spectrum allocations,” according to a news release from NYU Tandon School of Engineering.

 

Growing use of neurobiological evidence in criminal trials, new study finds

Science, ScienceInsider


from January 21, 2016

In 2008, in El Cajon, California, 30-year-old John Nicholas Gunther bludgeoned his mother to death with a metal pipe, and then stole $1378 in cash, her credit cards, a DVD/VCR player, and some prescription painkillers. At trial, Gunther admitted to the killing, but argued that his conviction should be reduced to second-degree murder because he had not acted with premeditation. A clinical psychologist and neuropsychologist testified that two previous head traumas—one the result of an assault, the other from a drug overdose—had damaged his brain’s frontal lobes, potentially reducing Gunther’s ability to plan the murder, and causing him to act impulsively. The jury didn’t buy Gunther’s defense, however; based on other evidence, such as the fact that Gunther had previously talked about killing his mother with friends, the court concluded that he was guilty of first-degree murder, and gave him a 25-years-to-life prison sentence.

Gunther’s case represents a growing trend, a new analysis suggests. Between 2005 and 2012, more than 1585 U.S. published judicial opinions describe the use of neurobiological evidence by criminal defendants to shore up their defense, according to a study published last week in the Journal of Law and the Biosciences by legal scholar Nita Farahany of Duke University in Durham, North Carolina, and colleagues. In 2012 alone, for example, more than 250 opinions cited defendants’ arguments that their “brains made them do it”—more than double the number of similar claims made in 2007.

 

How Larry Page’s Obsessions Became Google’s Business – The New York Times

The New York Times


from January 22, 2016

… Mr. Page is hardly the first Silicon Valley chief with a case of intellectual wanderlust, but unlike most of his peers, he has invested far beyond his company’s core business and in many ways has made it a reflection of his personal fascinations.

He intends to push even further with Alphabet, a holding company that separates Google’s various cash-rich advertising businesses from the list of speculative projects like self-driving cars that capture the imagination but do not make much money. Alphabet companies and investments span disciplines from biotechnology to energy generation to space travel to artificial intelligence to urban planning.

 

Human brain networks function in connectome-specific harmonic waves

Nature Communications


from January 21, 2016

A key characteristic of human brain activity is coherent, spatially distributed oscillations forming behaviour-dependent brain networks. However, a fundamental principle underlying these networks remains unknown. Here we report that functional networks of the human brain are predicted by harmonic patterns, ubiquitous throughout nature, steered by the anatomy of the human cerebral cortex, the human connectome. We introduce a new technique extending the Fourier basis to the human connectome. In this new frequency-specific representation of cortical activity, that we call ‘connectome harmonics’, oscillatory networks of the human brain at rest match harmonic wave patterns of certain frequencies. We demonstrate a neural mechanism behind the self-organization of connectome harmonics with a continuous neural field model of excitatory–inhibitory interactions on the connectome. Remarkably, the critical relation between the neural field patterns and the delicate excitation–inhibition balance fits the neurophysiological changes observed during the loss and recovery of consciousness.

 

Can Big Genomic Data Reveal the Fundamental Units of the Brain?

Bio-IT World


from January 20, 2016

An adult mouse’s brain, an object not much bigger than the last joint of your pinky finger, contains around 75 million neurons. At the Allen Institute for Brain Science in Seattle, the Mouse Cell Types program, led by Hongkui Zeng, is trying to figure out just how many varieties of neurons make up this vast complex, and what makes each one unique.

 

Weather Predictions Are Getting Better Thanks To New, Tiny Satellites

Fast Company, Co.Exist


from January 20, 2016

The consequences of bad weather forecasting are often expensive and dangerous. Not knowing until the last minute that a bad storm is going to arrive can leave homes and businesses unprepared. Or, a city may end up spending to prepare for a massive weather-related disruption that never actually materializes.

Sometimes bad weather forecasting is the product of bad forecasters or computer models. But other times it comes down to poor or incomplete data. And that’s where Spire, a startup that is launching the world’s first private weather satellite network, wants to step in.

 

Research gets increasingly international

Nature News & Comment


from January 19, 2016

China’s share of global science and engineering publications has pulled within a percentage point of those from the United States, according to the latest research statistics published by the US National Science Foundation (NSF).

The agency’s report, released on 19 January, also underscores the rising importance of international scientific collaboration. Between 2000 and 2013, the percentage of publications with authors from multiple countries rose from 13.2% to 19.2%.

 

MIT Sloan launches three new undergraduate majors and minors

MIT Sloan School of Management


from January 10, 2016

MIT Sloan undergraduate education, commonly known as Course 15, is about to undergo extensive restructuring for the first time in decades.

Starting this fall, Course 15 will offer three new majors—management, business analytics, and finance—and corresponding minors. Students in the classes of 2017 and 2018 can declare one of these new majors, or continue with the current management science major. Students in the classes of 2019 and beyond will choose one of the new majors.

 
Tools & Resources



Paper craft : Naturejobs

Naturejobs, Column


from January 20, 2016

Most scientific manuscripts today have anywhere from two to thousands of co-authors. Large collaborations — generally defined as more than a couple of dozen members — usually establish formal structures to govern manuscript writing, but smaller groups often develop their writing rules in an ad hoc way. As a result, we have seen — and participated in — countless small collaborations that struggle with such fundamental issues as editing their manuscript and deciding on the author list. Here we offer some general rules for small groups on how best to craft a scientific manuscript.

 

Organizing My Emails With A Neural Net

Andrey Kurenkov's Web World blog


from January 13, 2016

One of my favorite small projects, EmailFiler, was motivated by a school assignment for Georgia Tech’s Intro to Machine Learning class. Basically, the assignment was to pick some datasets, throw a bunch of supervised learning algorithms at them, and analyze the results. But here’s the thing: we could make our own datasets if we so chose. And so choose I did – to export my gmail data and explore the feasibility of machine-learned email categorization.

 

Leave a Comment

Your email address will not be published.