Daily Archives: February 6, 2017

SIPA Asia via ZUMA Wire

Atrios today:

Self-Checkouts

Those still a thing? I mean, I know they are, but around me the 3 major supermarkets within walking distance got rid of them....Anyway, I know they still exist, but I do think our robot future is not quite as inevitable as people think. Worrying about the impact of future automation on jobs seems to be a cool tech away of ignoring the current fucked and bullshit jobs situation. And, yes, automation has been going on for decades, which is actually my point. There's nothing new about it, and I don't know why people think there will be this sudden automation discontinuity. The robots have been here for awhile, and they aren't really going away, but that doesn't mean the sci-fi dystopian workless future is just around the corner. Shit is fucked up and bullshit enough without worrying about things which haven't happened yet, and likely won't.

It really doesn't matter if artificial intelligence is distracting us from whatever you think the "real" problem is. It's coming anyway. The speed of the AI revolution depends solely on fundamental factors (mostly continued reductions in the cost of parallel computing power) and the level of interest in AI software development. The fundamental factors are obviously still barreling ahead, and it sure looks like the free market has a ton of interest too:

Besides, AI is the real problem. As we all know (don't we?), the decline of manufacturing in the US has far more to do with automation than with trade or globalization. That decline set up the conditions for an angry working class in three northwestern states that finally decided it had found a savior in a guy who claimed it was all the fault of a bunch of foreigners. So now Donald Trump is president. How much more real can you get?

And that was just old-fashioned dumb automation. Smart automation is going to have a far bigger and far faster effect. We're not very far off from the first real destruction of an industry (probably long-haul trucking) thanks to smart automation, and after that it's going to come thick and fast.

So what are we going to do? Will our future be in the hands of demagogues who gain power by lashing out at scapegoats while they work hard to make sure that rich people get all the benefits of AI? Or will it be in the hands of people who actually give a damn about the working class and understand that a world of increasing automation requires a dramatic rethink of basic economics? I would sure like it to be the latter.

Unfortunately, like global warming, the effects of AI are slow and invisibleon a human timescale anyway. So it's easy to pretendno matter how idiotic this isthat AI is just a rerun of the Industrial Revolution. It's easy to pretend that each new advance isn't really a step toward true AI. It's easy to pretend that each individual industry to fall is just a special case. It's easy to pretend that something else is always more important.

Is AI coming soon? I find this question too boring to spend much time on anymore. Of course it's coming soon. The only question I'm interested in is what we're going to do about it. I keep pondering this, and I keep failing to come up with any likely answers that are very optimistic in the medium term. Maybe I'm not thinking outside the box enough. But it sure looks like we're determined to keep our collective heads in the sand for a long time. At best, the result is going to be a grim future of plutocracy for some and the dole for everyone else. At worst, it's going to be a future of global genocide (do you think there's enough aid in the world to keep Bangladesh afloat when there's no longer any work there?).

Eventually everything will work out, probably after a lot of suffering and a popular revolt. But wouldn't it be nice to avoid all that?

Oh, and those self-checkout machines? I don't know about Philly, but there's hardly a supermarket within ten miles of me that doesn't have them. Not only are they still a thing, but they're only going to get better. So sorry about all those nice union jobs as checkers and baggers.

Here is the original post:

Artificial Intelligence Is Coming Whether You Like It Or Not - Mother Jones

The Quebecois city of Montreal has long been known as a hotbed of creativity -- home of Cirque du Soleil and a hub for companies in the online gaming and special effects industries, not to mention its place as a financial and trade capital.

Creativity played a key role when the city of 2 million (with 4 million regionally) competed against other municipalities globally to win the 2016 title of Intelligent Community of the Year.

And now that commitment to creativity is spurring the city to explore a range of unique new smartphone apps and other startup-generated initiatives that leverage sensors, data collection and analysis, and machine learning to deal with snow removal, ever-increasing traffic and other municipal challenges.

Public Wi-Fi, smart mobility and digital public services are just some of the 70 municipal projects detailed in the city's Smart and Digital City Action Plan, begun in 2015. More than half of the projects are expected to be finished by 2018, though some will take longer.

"Montreal is known as the place 'where Shakespeare meets Moliere.' It's a creativity hub," says Harout Chitilian, the elected official in charge of the city's smart city initiatives and technology. "All these things meshing together make Montreal one of the greatest startup digital ecosystems."

By intent, the government has made that startup ecosystem a key compontent of its smart city push, says Chitilian, who serves as vice president of the city's executive committee, the executive branch of the municipal government that includes Mayor Denis Coderre.

Of the dozens of initiatives currently underway in Montreal, several involve partnerships with the private sector in which the city, Quebec Province and businesses share costs. Those projects range from a high-speed, fiber-optic Scientific Information Network to eight different smart mobility and parking projects.

The principal driver of this partnership is InnoCit MTL, an independent, non-profit tech accelerator that receives both city and business financial support. Housed in the historic Notman House in downtown Montreal, InnoCit MTL has already fostered more than 15 startups in just over a year.

Notman House was alive with activity when Computerworld visited during a cold snap in mid-December, 2016 as part of a three-day tour of this smart city. Here's what we found.

The city government, along with the Province of Quebec and members of the academic community, have put special focus on artificial intelligence. Those efforts meld well with private sector startups that likewise are tapping the power of AI.

One such startup is Infra.AI, which intends to use machine learning and artificial intelligence to scan high resolution images of the city's streets and buildings."The benefits of AI are numerous," says co-founder Franois Maillet. "The fact that Montreal is serious about smart city and investing in it, there's a direct and positive impact in the startup community and the R&D. For the city itself, it provides better services to the citizens."

LIDAR images can help municipalities like Montreal monitor city infrastructure to identify such changes in status as detoriating bridges, broken windows or building code violations.

With digital image information from satellites, low-flying planes and LIDAR-equipped city vehicles, technology under development at Infra.AI will make it possible for Montreal and other cities to provide almost-real-time data on street conditions or the safety of roads and bridges.

That data can be combined with information from traffic video sensors and sensors on buildings, says Maillet, who also co-founded a related startup, MLDB.AI, that is working on a machine-learning database.

The potential applications are far-ranging. A firetruck speeding to a fire might be automatically advised that there's an obstruction in the roadway, allowing it to take another pathway. Or a pothole larger than a foot could be spotted, automatically dispatching a road crew to patch it. AI can even help identify a sagging highway bridge span, noticing a small drop when compared with the previous scans from days or weeks earlier.

Montreal-based Infra.AI is employing pattern recognition intelligence to distinguish a group of pedestrians from vehicles. The software could be used to identify problem locations and develop systems for improved pedestrian safety.

Infra.AI is currently piloting a program that helps identify ailing trees on city streets, a problem plaguing Montreal right now. When the startup's AI system is shown images of healthy trees, it can compare those with recent imagery to identify less-healthy trees with patches and browning leaves that need to be maintained or replaced.

"When you think of the kind of data [already] coming in from LIDAR and cameras, it's huge. The applications are now becoming possible with AI," says Jean-Franois Gagn, CEO of Element AI, a Montreal-based incubator dedicated to matching AI startups with larger companies and with government agencies.

Through its Canada First Research Excellence Fund, the Canadian government last year provided about $200 (US) million to three Montreal-based universities for research that Gagn believes will yield sophisticated AI spinoff companies in 2017.

In addtion, both Google and Microsoft have recently made investments in Montreal-based AI.

On a more personal level, another InnoCit MTL startup, Key2Access, is getting ready to test an app to make it safer for disabled people to cross city streets, according to CEO Sophie Aladas. Key2Access's tech is already being piloted in Ottawa, and has been successfully tested there by Richard Marsolais, a man with a vision impairment who is a specialist in independent living for the Canadian National Institute for the Blind.

Marsolais and his guide dog, Ashland, along with Motaz Aladas, head engineer for Key2Access (and CEO Sophie's father), demonstrated for Computerworld at a Montreal intersection how a small handheld device or a smartphone could be used to activate a Bluetooth-enabled crosswalk signal, making it safe for a vision-impaired or disabled person to cross. (See Smart Cities: Montreal for video footage of that demonstration.)

Marsolais says it would be helpful to have a handheld activation device to change the signal, instead of relying only on his guide dog or an audible crossing signal, which isn't always easy to hear. In addition, it isn't always clear in which direction it's safe to cross; Key2Access aims to solve that problem by using audible commands or vibrations to direct the user onto the crosswalk in the proper direction.

For Key2Access to function, traffic engineers in Montreal will need to install a receiver at each intersection to receive the wireless signal from the handheld device, Aladas says. The cost will be comparable to enabling a traditional crosswalk button on a pole, Sophie Aladas says. The city is expected to install the gear on at least one intersection in the spring as part of the testing phase.

A number of initiatives are in the works to help reduce traffic in Montreal in the next two years, including a tripling of the number of intelligent traffic signals to reach 2,200 units.

Data from the 700 existing smart signals installed over the last two years and from 500 surveillance cameras and Bluetooth sensors already helps prioritize buses traveling the streets to lessen commute times by 15% to 20%, the city's Chitilian says, with more improvements expected. Montreal is also in partnership with Waze, Google's crowdsourcing traffic app, to help syphon off driver data for greater intelligence.

In addition to its efforts to lessen traffic congestion and improve the efficiency of public transportation, Montreal heavily promotes bicycle riding. It's not uncommon to see bicyclists pedaling through downtown streets even in the dead of winter.

Bixi, a bike-sharing system, got its start in Montreal in 2009; as of 2015, there were 3.5 million Bixi rides each year in the city, and the service has grown to 45,000 bikes in 15 cities. The Bixi mobile apps for iOS and Android, along with other Bixi add-ins developed by Montreal startups, allow everything from online payments to personal fitness tracking for the bikes.

Separately, Montreal startup SmartHalo is testing technology to turn any bike into a smart bike using a rider's smartphone and its GPS connection.

"We know for a fact that adding preferential lights and dedicated bus lanes increases the speed of going from point A to B and makes the service much more efficient. You can have the same amount of buses and workable hours with better service," Chitilian says.

Sensor data from traffic signals is already being sent to a recently created central command post -- a "decision center," Chitilian calls it -- where technicians pore over dozens of desktop monitors and large wall displays. "The center gives us the ability to have an overall view" of the city, helping if there is an accident or other public safety need, he says.

Montreal also has designated $76 million US to replace 100,000 streetlights in the next five years with more efficient LED lighting that will be equipped with sensor and communications technology to expand the city's ability to manage congestion, pedestrian crowds, accidents and more, according to Chitilian.

With its combination of AI-focused startup innovation, sensor-driven traffic-improvement initiatives and data-driven apps for citizen empowerment, Montreal seems well on its way to furthering its designation as an intelligent city.

"We are trying to build a smart city from the ground up, and are putting in the pillars to do it," Chitilian says. "As politicians, we have to show immediate results, but some of our decisions will have lasting impact beyond our political mandates," he muses.

"We have to make decisions that will look good down the road," Chitilian says. "What we have in Montreal is more than optimism. It is a generational transformation."

Montreal and the Quebec Province have committed to sharingpublicly available data, which private enterpreneurs have put to innovative use via smartphone apps. Here are a few of locals' favorites:

More:

Montreal sees its future in smart sensors, artificial intelligence (with video) - Computerworld

Babak Hodjat believes humans are too emotional for thestock market. So he's started one of the first hedge funds run completely by artificial intelligence.

"Humans have bias and sensitivities, conscious and unconscious," says Hodjat, a computer scientist who helped laythe groundwork for Apple's Siri. "It's well documented we humans make mistakes. For me, it's scarier to be relying on those human-based intuitions and justifications than relying on purely what the data and statistics are telling you."

Babak Hodjat

Photographer: David Paul Morris/Bloomberg

Hodjat, with 21 patents to his name, is co-founder and top scientist of Sentient Technologies Inc., a startup that has spent nearly a decadelargely in secrettraining an AI system that can scour billions of pieces of data, spot trends, adapt as it learns and make money trading stocks. The team of technology-industry vets is betting that softwareresponsible forteaching computers to drive cars, beat the world's best poker players and translate languages will give their hedge fund an edge on Wall Street pros.

The walls of Sentient's San Francisco office are dotted with posters for robots-come-alive movies such as "Terminator." Inside a small windowless trading room, the only light emanates fromcomputer screens and a virtual fire on a big-screen TV. Two guys are quietly monitoring the machine's tradesjust in case the system needs to be shut down.

If all hell breaks loose," Hodjat says, "there is a red button."

Sentient won't disclose its performance or many details about the technology, and the jury is out on the wisdom of handing off trading to a machine. While traditional hedge funds including Bridgewater Associates, Point72 and Renaissance Technologies have poured money into advanced technology, many use artificial intelligence to generate ideasnot to control their entire trading operations.

All the same, Sentient, which currently trades only its own money, is being closely watched by the finance and AIcommunities. The venture capital firm owned by Hong Kong's richest man, Li Ka-shing, and India's biggest conglomerate, Tata Group, are among backers who have given the company $143 million. (Beyond trading, Sentient's AI system is being applied to a separate e-commerce product.)

Trading is "one of the top 10 places that AI can make a difference," says Nello Cristianini, a professor of artificial intelligence at the University of Bristol who has been advising Sentient. "A trading algorithm can look at the data, make a decision, act and repeatyou can have full autonomy."

Sentient's team includes veterans of Amazon, Apple, Google, Microsoft and other technology companies. They're part of a small group in Silicon Valley using expertise in data science and the field of artificial intelligence known as machine learning to try and disrupt financial markets.

AI scientists typically have no interest in working for a hedge fund, says Richard Craib, who started the AI hedge fund Numerai. "But they may want to mess around with data sets." Numerai's system makes trades by aggregating trading algorithms submitted by anonymous contributors who participate in a weekly tournament where prizes are awarded in Bitcoin. It recently raised $6 million from investors including Howard Morgan, the co-founder of the quant investment management firm Renaissance Technologies. "It's entirely a data science problem," Craib says.

Another company, called Emma, started a hedge fund last year based on an artificial intelligence system that can write news articles.

Employees of Sentient Technologies in San Francisco.

Photographer: David Paul Morris/Bloomberg

Hodjat of Sentient spent much of his career focused on the language-detection technology behind smartphone digital assistants. Several employees from his previous company, Dejima, went on to create Apple's Siri. Rather than join, he chose to focus on advances in artificial intelligence. His career goals didn't include finance, but he sees markets as one of the most promising applications for the technology. The vast amounts of publicly available data, along with stronger computers to analyse it for patterns, make the field an ideal fit. "That is the fuel for AI," he says.

Sentient's system is inspired by evolution. According to patents, Sentient has thousands of machines running simultaneously around the world, algorithmically creating what are essentially trillions of virtual traders that it calls "genes." These genes are tested by giving them hypothetical sums of money to trade in simulated situations created from historical data. The genes that are unsuccessful die off, while those that make money are spliced together with others to create the next generation. Thanks to increases in computing power, Sentient can squeeze 1,800 simulated trading days into a few minutes.

An acceptable trading gene takes a few days and then is used inlive trading. Employees set goals such as returns to achieve, risk level and time horizon, and then let the machines go to work. The AI system evolves autonomously as it gains more experiences.

Sentient typically owns a wide-ranging batch of U.S. stocks, trading hundreds of times per day and holding positions for days or weeks. "We didn't impose that on the system," says Jeff Holman, the company's chief investment officer. "The artificial intelligence seems to agree with what you get from human intelligence that it's better to spread your bets and have a more diversified portfolio."

As impressive as Sentient's technology appears, it's hard to know if it works. The company says the AI system is beating internal benchmarks, but won't disclose what those are. It shares little about the data used for the AI's decision-makingand isn't profitable. The company plans to bring in outside investors later this year. Holman, a Wall Street veteran who joined last year, said thecompany is limited on what it can say by U.S. Securities Exchange Commission rules restricting marketing by hedge funds that are raising money. "The platform is solid," he says. "It doesn't look like any other strategy I've seen."

Anthony Ledford, the chief scientist at the $19 billion hedge fund Man AHL in London, warns of putting too much faith in this branch of artificial intelligence without more evidence. Man AHL uses machine learning for a portion of its clients money, and Ledford is encouraged by the results. While the company is exploring a standalone machine-learning strategy, he says it's too early to declare success."There's a lot of hype and promise," Ledford says. "But when you actually ask people how many hundreds of millions dollars they are trading, many of them don't come back with much at all."

Little performance data is available about AI-focused hedge funds. One index that tracks 12 pools that utilize AI as part of its core strategies, called Eurekahedge AI Hedge Fund Index, returned 5 percent last year. That's slightly better than the average hedge fund, but trailed the S&P 500.

Tristan Fletcher, who wrote his doctoral thesis on machine learning in financial markets and works for a hedge fund, says investors may be reluctant to turn over their money completely to a machine. "I know how conservative investors are and I know of no one who would put their money in asystem that's fully systematic," says Fletcher. "Machine learning isn't a panacea for everything. You need people who have literal thinking."

Exclusive insights on technology around the world.

Get Fully Charged, from Bloomberg Technology.

with Nishant Kumar in London

Excerpt from:

Silicon Valley Hedge Fund Takes On Wall Street With AI Trader - Bloomberg

An artificial intelligence called Libratus beats four of the worlds best poker players in Pittsburgh last week. Photograph: Carnegie Mellon University

First it was checkers (draughts to you and me), then chess, then Jeopardy!, then Go and now poker. One after another, these games, all of which require significant amounts of intelligence and expertise if they are to be played well, have fallen to the technology we call artificial intelligence (AI). And as each of these milestones is passed, speculation about the prospect of superintelligence (the attainment by machines of human-level capabilities) reaches a new high before the media caravan moves on to its next obsession du jour. Never mind that most leaders in the field regard the prospect of being supplanted by super-machines as exceedingly distant (one has famously observed that he is more concerned about the dangers of overpopulation on Mars): the solipsism of human nature means that even the most distant or implausible threat to our uniqueness as a species bothers us.

The public obsession with the existential risks of artificial superintelligence is, however, useful to the tech industry because it distracts attention from the type of AI that is now part of its core business. This is weak AI and is a combination of big data and machine-learning algorithms that ingest huge volumes of data and extract patterns and actionable predictions from them. This technology is already ubiquitous in the search engines and apps we all use every day. And the trend is accelerating: the near-term strategy of every major technology company can currently be summarised as AI Everywhere.

The big data/machine-learning combination is powerful and enticing. It can and often does lead to the development of more useful products and services search engines that can make intelligent guesses about what the user is trying to find, movies or products that might be of interest, sources of information that one might sample, connections that one might make and so on. It also enables corporations and organisations to improve efficiency, performance and services by learning from the huge troves of data that they routinely collect but until recently rarely analysed.

Human freedoms and options are increasingly influenced by opaque, inscrutable algorithms

Theres no question that this is a powerful and important new technology and it has triggered a gadarene stampede of venture and corporate capital. We are moving into what one distinguished legal scholar calls the black box society, a world in which human freedoms and options are increasingly influenced by opaque, inscrutable algorithms. Whose names appear on no-fly lists? Who gets a loan or a mortgage? Which prisoners get considered for parole? Which categories of fake news appear in your news feed? What price does Ryanair quote you for that particular flight? Why has your credit rating suddenly and inexplicably worsened?

In many cases, it may be that these decisions are rational and/or defensible. The trouble is that we have no way of knowing. And yet the black boxes that yield such outcomes are not inscrutable to everyone just to those who are affected by them. They are perfectly intelligible to the corporations that created and operate them. This means that the move towards an algorithmically driven society also represents a radical power-shift, away from citizens and consumers and towards a smallish number of powerful, pathologically secretive technology companies, whose governing philosophy seems to be that they should know everything about us, but that we should know as little as possible about their operations.

Whats even more remarkable is that these corporations are now among the worlds largest and most valuable enterprises. Yet, on the whole, they dont receive the critical scrutiny their global importance warrants. On the contrary, they get an easier ride from the media than comparable companies in other industries. If the CEO of an oil company, a car manufacturer or a mining corporation were to declare, for example, that his motto was Dont Be Evil, even the most somnolent journalist might raise a sceptical eyebrow. But when some designer-stubbled CEO in a hoodie proclaims his belief in the fundamental goodness of humanity, the media yawn tolerantly and omit to notice his companys marked talent for tax avoidance. This has to stop: transparency is a two-way process.

See more here:

The Observer view on artificial intelligence - The Guardian

New research indicates that Artificial Intelligence, or AI, as it is defined and practised today, has several limits. New buzzwords only serve to mystify the populace, and it is increasingly clear to me that many technologists and information technology (IT) managers are just groping about in the dark. They throw out terms such as neural networks, deep learning, big data, black box systems, and so on, hoping to mask the fact that they know very little of how this technology may evolve over the next several years.

As an observer, I cant help but think there is an important question in front of us: are the ramblings of these pundits in fact a case of the one-eyed man becoming king in the land of the blindor, instead, more akin to the parable of the five blind men, who all encountered an elephant and, after inspecting various parts of the elephant by touch, came away with different definitions of what an elephant is like?

The vital premise in todays AI is that the computer program itself learns as it goes along, creating a database of information, and then, uses that database to automatically generate additional computer programming codes as it learns morewithout the need for human programmers. These AI programs then become black boxes, since even their original human programmers have no way of knowing what code the machine has generated on its own.

ALSO READ: The road ahead for AI: engendering trust

These computer programs, however, need copious amounts of carefully categorized data to make themselves smarter. Anything that is sloppily characterized can easily cause the machine to make the wrong conclusions. I have mentioned before in this column that it has been proven that just changing a few pixels on an image can make an AI image-recognition program conclude that a car is in fact an elephantwhich is a mistake that an ordinarily intelligent human eye would never make.

Thus, many firms that are trying to chart out a path in AI are scrambling to go out and acquire vast stores of data that have already been neatly characterized. IBM, for instance, has bought firms that own billions of medico-radiological imagesin the hope of feeding this vast acquired data to the medical diagnosis components of IBMs Watson product. The idea is that this data, collected over many years of digital medico-radiological imaging, will enable Watson to become cannier in diagnosing diseases. When quizzed about these acquisitions, a senior IBM executive said to me recently: If youre not at the table, you can be sure youll be on the menu.

In another example of the use of categorized data, a firm called Cambridge Analytica has recreated a sinister way to profile people, from psychometric tests that show up, ostensibly as harmless quizzes, on Facebook and other social networking sitesluring people into taking them and posting the individual results online. Cambridge Analytica claims it used these psychometric analyses to accurately predict the personality types and preferences of individual voters. The firm was apparently retained by both the Brexit leave and Donald Trumps presidential election campaigns to accurately target voters who were likely to vote for them, and to lure more of these supportive voters out to the polling booths.

Trained psychologists have a dim view of psychometric testing and other personality profiling tests. When I asked my sister, who holds a doctorate from Harvard in Psychology, about the efficacy of such methods, her response was that there are dozens of such psychometric rubrics out there that do have some utility, but are in fact quite flawed; many of them have been debunked for predictive utility.

The accuracy of diagnostics and psychometrics aside, the fact remains that without reams of carefully categorized data, AI as we know it today is dead on arrival. That means that in areas where data is not yet availablefor instance, crash data for self-driving carswe must look elsewhere to create models that mimic large data stores accurately when data is absent. Where does one go to find out under what circumstances self-driving automobiles like the Tesla that killed its occupant in 2016 might have other such accidents? Enough instances of this havent occurred and, therefore, the data doesnt exist. Building predictive models here without data is not neuralits neurotic, and dangerous!

ALSO READ: Why India needs an AI policy

This brings us to the fields of pure mathematics and theoretical physics, which are the way forward. In an informative blog last year, Wale Akinfaderin, a Ph.D. candidate in physics at Florida State University, has enumerated the types of mathematics that an aspiring AI specialist must be familiar with, if not master, to be effective. Here is a partial list from his blog post: Principal Component Analysis, Eigen decomposition, Combinatorics, Bernoulli, Gaussian, Hessian, Jacobian, Laplacian, and Lagragian Distributions, Entropy, and Manifolds. Ill stop hereIm sure you get the idea!

Dont panic, says Neil Sheffield, an AI researcher at Amazon, in a blog. By bringing our mathematical tools to bear on the new wave of deep learning methods, we can ensure they remain mostly harmless.

Time for us amateur pundits and pedestrian programmers to make way for the pure mathematicians and theoretical physicists to lead the charge. They have long used mathematical theory to contemplate the unsolvable where data doesnt exist. Visionaries like Stephen Hawking, Albert Einstein and Srinivasa Ramanujan have been feted for their ability to posit plausible models on hitherto unsolvable problems such as the theory of the universe.

One-eyed they may well be, but all hail the new kings of AI!

Siddharth Pai is a world-renowned technology consultant who has led over $20 billion in complex, first-of-a-kind outsourcing transactions.

First Published: Tue, Feb 07 2017. 12 58 AM IST

Originally posted here:

Allow mathematicians to pierce artificial intelligence frontiers - Livemint