Artificial Intelligence Development


Likely author of original Bitcoin paper? Researchers uncover linguistic evidence

The primary author of the celebrated Bitcoin paper is most likely Nick Szabo, a blogger and former George Washington University law professor, according to students and researchers at Aston University's Centre for Forensic Linguistics.

Their research also challenges the claim that Dorian S. Nakamoto is the main author of the paper, an assertion made in Newsweek last month that has been strongly denied by Mr Nakamoto.

The study, nicknamed 'Project Bitcoin', was undertaken by a team of 40 final-year forensic linguistics students led by Dr Jack Grieve, Lecturer in Forensic Linguistics at Aston University. It looked at linguistic similarities between the Bitcoin cryptocurrency paper and the writing of eleven other individuals that are regularly referred to as possible authors: Dorian S. Nakamoto, Vili Lehdonvirta, Michael Clear, Shinichi Mochizuki, Gavin Andresen, Nick Szabo, Jed McCaleb, Dustin D. Trammel, Hal Finney, Wei Dai, and Neal King, Vladimir Oksman & Charles Bry.

Bitcoin is an internet-based virtual currency which allows users to buy goods and services online. The payment system, introduced in 2009, is supposedly easier and safer than sending money via more traditional means. Using Bitcoin to pay for items also means avoiding credit card, foreign exchange or cash handling fees.

Dr Grieve said: "The number of linguistic similarities between Szabo's writing and the Bitcoin paper is uncanny, none of the other possible authors were anywhere near as good of a match. We are pretty confident that out of the list of people regularly referred to as possibilities, Nick Szabo is the main author of the paper, though we can't rule out the possibility that others contributed.

"Our study adds to the weight of evidence pointing towards Nick Szabo. The case looks pretty clear-cut. Szabo is an expert in law, finance, cryptography and computer science. He created 'bit gold', a precursor to Bitcoin, and was looking for collaborators in 2008. Did Nick Szabo create Bitcoin? We're not sure, but we think he probably wrote the paper so it's certainly worth a closer look."

The results showed that of the eleven Szabo is by far the closest match, with a large number of distinctive linguistic traits appearing in both the Bitcoin paper and Szabo's blogs and other writings. This includes the use of: the phrases "chain of…," "trusted third parties," "for our purposes," "need for…," "still," "of course," "as long as," "such as" and "only" numerous times, contractions, commas before 'and' and 'but', hyphenation, '-ly' adverbs, the pronouns 'we' and 'our' in papers by a single author; fragmented sentences following colons and reflexive (-self) pronouns.

In total hundreds of documents written by the eleven possible authors were considered, including over 40 academic papers written by Szabo which are available on his personal website.

The study also questioned why the most-cited textual feature of the Bitcoin paper is the fact that it contains double spaces after full stops. The Bitcoin paper was drafted using Latex, an open source document preparation system. Without the base .tex for the Bitcoin paper, which is not available, researchers are unable to tell if the author double spaces between sentences. However, the study noted that Szabo uses Latex for all his publications.

Story Source:

The above story is based on materials provided by Aston University. Note: Materials may be edited for content and length.


It's a bubble, but not as we know it

Multi-sensory technology that creates soap bubbles, which can have images projected onto them or when the bubbles are burst release a scent, will be unveiled at an international conference later this month.

The research paper, to be presented at one of the world's most important conferences on human-computer interfaces -- ACM CHI 2014 [26 April-1 May], could be used in areas such as gaming or education and encourage a new way of thinking about multi-sensory technologies.

SensaBubble, led by Professor Sriram Subramanian from the University of Bristol's Department of Computer Science, is a chrono-sensory mid-air display system that generates scented bubbles to deliver information to people using different senses.

The bubble-based technology creates bubbles with a specified size and frequency, fills them with an opaque fog that is optionally scented, controls their route, tracks their location and projects an image onto them.

SensaBubble uses the concept of chrono-sensory experiences where layers of information are presented via different senses for variable length of times, each attracting different types of interest from the user. Firstly, a visual display projected onto the bubble which only lasts until it bursts; secondly, a scent released upon the bursting of the bubble slowly disperses and leaves a longer-lasting noticeable trace.

Sriram Subramanian, Professor of Human-Computer Interaction in the University's Bristol Interaction and Graphics group, said: "The human sense of smell is powerful, but there are few research systems that explore and examine ways to use it. We have taken the first steps to explore how smell can be used to enhance and last longer in a visual object such as a soap bubble.

"There are many areas in which bubble-based technology like SensaBubble could be applied, such as a SensaBubble clock that releases the number of scented bubbles corresponding to the hour or SensaBubble Maths, an educational game for children, which incorporates smell as feedback on their success."

Interactive technologies that are directly targeted at generating public interest and drawing the user's attention have many applications in advertising and certain forms of education, such as museum exhibits.

SensaBubble could be adapted for use in a variety of areas including education, alerts and engaging user experiences.

A video of SensaBubble is available on YouTube at:

Paper: SensaBubble: a chrono-sensory mid-air display of sight and smell, Sue Ann Seah, Diego Martinez Plasencia, Peter Bennett, Abhijit Karnik, Vlad Otrocol, Jarrod Knibbe, Andy Cockburn, and Sriram Subramanian, Proceedings of ACM CHI 2014 Conference on Human Factors in Computing Systems. Toronto, Canada. 2014., 26 April-1 May 2014.

Story Source:

The above story is based on materials provided by University of Bristol. Note: Materials may be edited for content and length.


Superconducting qubit array points the way to quantum computers

A fully functional quantum computer is one of the holy grails of physics. Unlike conventional computers, the quantum version uses qubits (quantum bits), which make direct use of the multiple states of quantum phenomena. When realized, a quantum computer will be millions of times more powerful at certain computations than today's supercomputers.

A group of UC Santa Barbara physicists has moved one step closer to making a quantum computer a reality by demonstrating a new level of reliability in a five-qubit array. Their findings appear Thursday in the journal Nature.

Quantum computing is anything but simple. It relies on aspects of quantum mechanics such as superposition. This notion holds that any physical object, such as an atom or electron -- what quantum computers use to store information -- can exist in all of its theoretical states simultaneously. This could take parallel computing to new heights.

"Quantum hardware is very, very unreliable compared to classical hardware," says Austin Fowler, a staff scientist in the physics department, whose theoretical work inspired the experiments of the Martinis Group. "Even the best state-of-the-art hardware is unreliable. Our paper shows that for the first time reliability has been reached."

While the Martinis Group has shown logic operations at the threshold, the array must operate below the threshold to provide an acceptable margin of error. "Qubits are faulty, so error correction is necessary," said graduate student and co-lead author Julian Kelly who worked on the five-qubit array.

"We need to improve and we would like to scale up to larger systems," said lead author Rami Barends, a postdoctoral fellow with the group. "The intrinsic physics of control and coupling won't have to change but the engineering around it is going to be a big challenge."

The unique configuration of the group's array results from the flexibility of geometry at the superconductive level, which allowed the scientists to create cross-shaped qubits they named Xmons. Superconductivity results when certain materials are cooled to a critical level that removes electrical resistance and eliminates magnetic fields. The team chose to place five Xmons in a single row, with each qubit talking to its nearest neighbor, a simple but effective arrangement.

"Motivated by theoretical work, we started really thinking seriously about what we had to do to move forward," said John Martinis, a professor in UCSB's Department of Physics. "It took us a while to figure out how simple it was, and simple, in the end, was really the best."

"If you want to build a quantum computer, you need a two-dimensional array of such qubits, and the error rate should be below 1 percent," said Fowler. "If we can get one order of magnitude lower -- in the area of 10-3 or 1 in 1,000 for all our gates -- our qubits could become commercially viable. But there are more issues that need to be solved. There are more frequencies to worry about and it's certainly true that it's more complex. However, the physics is no different."

According to Martinis, it was Fowler's surface code that pointed the way, providing an architecture to put the qubits together in a certain way. "All of a sudden, we knew exactly what it was we wanted to build because of the surface code," Martinis said. "It took a lot of hard work to figure out how to piece the qubits together and control them properly. The amazing thing is that all of our hopes of how well it would work came true."

Story Source:

The above story is based on materials provided by University of California - Santa Barbara. The original article was written by Julie Cohen. Note: Materials may be edited for content and length.


Take notes by hand for better long-term comprehension

Dust off those Bic ballpoints and college-ruled notebooks: research shows that taking notes by hand is better than taking notes on a laptop for remembering conceptual information over the long term. The findings are published in Psychological Science, a journal of the Association for Psychological Science.

Walk into any university lecture hall and you're likely to see row upon row of students sitting behind glowing laptop screens. Laptops in class have been controversial, due mostly to the many opportunities for distraction that they provide (online shopping, browsing Reddit, or playing solitaire, just to name a few). But few studies have examined how effective laptops are for the students who diligently take notes.

"Our new findings suggest that even when laptops are used as intended -- and not for buying things on Amazon during class -- they may still be harming academic performance," says psychological scientist Pam Mueller of Princeton University, lead author of the study.

Mueller was prompted to investigate the question after her own experience of switching from laptop to pen and paper as a graduate teaching assistant:

"I felt like I'd gotten so much more out of the lecture that day," says Mueller, who was working with psychology researcher Daniel Oppenheimer at the time. "Danny said that he'd had a related experience in a faculty meeting: He was taking notes on his computer, and looked up and realized that he had no idea what the person was actually talking about."

Mueller and Oppenheimer, who is now at the UCLA Anderson School of Management, conducted a series of studies to investigate whether their intuitions about laptop and longhand note-taking were true.

In the first study, 65 college students watched one of five TED Talks covering topics that were interesting but not common knowledge. The students, who watched the talks in small groups, were either given laptops (disconnected from Internet) or notebooks, and were told to use whatever strategy they normally used to take notes.

The students then completed three distractor tasks, including a taxing working memory task. A full 30 minutes later, they had to answer factual-recall questions (e.g., "Approximately how many years ago did the Indus civilization exist?") and conceptual-application questions (e.g., "How do Japan and Sweden differ in their approaches to equality within their societies?") based on the lecture they had watched.

The results revealed that while the two types of note-takers performed equally well on questions that involved recalling facts, laptop note-takers performed significantly worse on the conceptual questions.

The notes from laptop users contained more words and more verbatim overlap with the lecture, compared to the notes that were written by hand. Overall, students who took more notes performed better, but so did those who had less verbatim overlap, suggesting that the benefit of having more content is canceled out by "mindless transcription."

"It may be that longhand note takers engage in more processing than laptop note takers, thus selecting more important information to include in their notes, which enables them to study this content more efficiently," the researchers write.

Surprisingly, the researchers saw similar results even when they explicitly instructed the students to avoid taking verbatim notes, suggesting that the urge to do so when typing is hard to overcome.

The researchers also found that longhand note takers still beat laptop note takers on recall one week later when participants were given a chance to review their notes before taking the recall test. Once again, the amount of verbatim overlap was associated with worse performance on conceptual items.

"I don't anticipate that we'll get a mass of people switching back to notebooks," says Mueller, "but there are several new stylus technologies out there, and those may be the way to go to have an electronic record of one's notes, while also having the benefit of being forced to process information as it comes in, rather than mindlessly transcribing it."

"Ultimately, the take-home message is that people should be more aware of how they are choosing to take notes, both in terms of the medium and the strategy," Mueller concludes.


Online retailers have clear advantage by not collecting sales tax

Two independent studies use two very different approaches to reach the same conclusion: some online retailers really do have an advantage over traditional brick-and-mortar stores.

The studies find evidence from investors, analysts and consumers themselves that suggest online stores have a competitive edge when they don't have to collect sales tax from shoppers.

Both studies were conducted by researchers at the Fisher College of Business at The Ohio State University and their colleagues.

In one study, Brian Baugh, Itzhak Ben-David, and Hoonsuk Park from Ohio State's Department of Finance found that sales fell 9.5 percent at in five states when the online retailer began collecting taxes on online purchases.

In the other study, Jeffrey Hoopes, assistant professor of accounting, and his colleagues found that the stock prices of online retailers dropped when news broke about possible legislation that would require them to collect sales tax.

Overall, the two studies show that consumers do indeed spend less at online retailers when they have to pay sales tax -- and investors are quite aware of the threat these firms face from new tax-collection laws.

"Internet taxation is an important issue that will be debated for years to come," Ben-David said.

"But we're starting to learn how much internet retailers really benefit from not having to collect sales tax from customers in some states."

In their study, posted on the Social Science Research Network and National Bureau of Economic Research, Baugh, Ben-David, and Park used anonymized data from a personal finance website.

The researchers examined retail transactions for about 1.3 million households in 2012 and 2013 from five states that began collecting tax on Amazon purchases during this time: Texas, Pennsylvania, California, New Jersey and Virginia.

They found that after the introduction of the tax collection, Amazon sales fell 9.5 percent among these consumers.

The effect was particularly pronounced on larger purchases. Consumers decreased their spending at Amazon by nearly 24 percent on purchases of $300 and above.

"If you're going to make a big-ticket purchase like a big-screen TV or a laptop, there are currently huge incentives to go online to avoid the sales tax. These incentives are much stronger for large purchases than for small purchases, and our findings confirm that large purchases are indeed more sensitive to the introduction of this tax," said Baugh, a doctoral student.

Added Park, who is also a doctoral student, "If that advantage disappears, like it did in these states, you're more likely to find another online retailer where you don't have to pay the tax or go to a local store where you can see the product and return it easier."

But if the goal of these new tax-collection laws was to help local brick-and-mortar stores compete against Amazon, they only had limited success.

"For the most part, consumers simply switched their spending to other online retailers that didn't have to collect the tax," said Ben-David, associate professor of finance.

The researchers estimate that households use local brick-and-mortar retailers for about half of the lost Amazon purchases. Other purchases are done online.

These findings show that, overall, the piecemeal state-by-state legislation to tax online purchases will only give slight benefits to local brick-and-mortar stores.

"Unless there is some kind of national legislation, there will be loopholes that consumers can use to continue to avoid sales tax on some online purchases," Park said.

"It is clear that one of Amazon's advantages in the competition against other retailers was lower total price since Amazon did not collect sales tax. Our study shows that about 10 percent of sales were driven by this advantage," Baugh said.

But the drop in sales at Amazon in these five states shows why investors and analysts are clearly worried about how sales-tax legislation may affect online firms that are forced to collect the tax.

The Fisher College's Hoopes, along with colleagues Jacob Thornock and Braden Williams of the University of Washington, showed this in their study, which is also posted on the Social Science Research Network.

They looked at proposed federal bills that, if passed, would force online retailers to begin collecting sales tax. They then identified major event dates for each piece of legislation in which there was news coverage suggesting that passage of the bills was more likely.

The question was, what happened to the stock values of these online retailers in the days following these news announcements?

Hoopes and his colleagues found the online retailers saw their stock returns drop 0.7 percent more than expected on days when these proposed laws hit the news.

"Investors perceive that these e-tailers have a competitive advantage because they don't have to collect sales tax, and when that advantage is threatened, their stock prices go down," Hoopes said.

However, the researchers did not find that other retailers -- those firms that are suspected to be at a disadvantage compared to online retailers -- had an abnormally high stock return on days when the proposed sales tax legislation was in the news, Hoopes said. Instead, their returns were just about what would have been expected on these days.

In addition, the study found that, around the times of these news events, stock analysts forecast about a 1.5 percent decline in projected future sales revenue for online retailers as a result of potential federal sales tax legislation.

Hoopes and his colleagues also used a case study to show how Amazon sales in particular may suffer in states where they have to collect sales tax.

In this case, the researchers compared changes in internet searches for the word "Amazon" surrounding dates when Amazon first began collecting sales tax in various states. The hypothesis is that consumers may search for the word "Amazon" to get to the firm's website when they want to buy something online.

Results showed that weekly Google search volume for the term "Amazon" decreases in a given state after Amazon begins collecting sales tax in that state.

Overall, Hoopes said this study provides a company-level view of the effects of sales-tax legislation in the states.

"Most of the previous research focuses on the consumer, and whether they will buy fewer products from e-tailers if they have to pay sales tax," Hoopes said. "We examined if collecting sales tax would actually hurt the firms themselves, and found evidence that it would.

"The concerns that there isn't a level playing field for all retailers may be justified, as least as it is perceived by the market."


Page 2 of 28

Our Partners