IBM's Watson Is Now Analyzing Your Vacation Photos 117

jfruh writes: IBM's Jeopardy-winning supercomputer Watson is now suite of cloud-based services that developers can use to add cognitive capabilities to applications, and one of its powers is visual analysis. Visual Insights analyzes images and videos posted to services like Twitter, Facebook and Instagram, then looks for patterns and trends in what people have been posting. Watson turns what it gleans into structured data, making it easier to load into a database and act upon — which is clearly appealing to marketers and just as clearly carries disturbing privacy implications.

An AI Hunts the Wild Animals Carrying Ebola 45

the_newsbeagle writes: Outbreaks of infectious diseases like Ebola follow a depressing pattern: People start to get sick, public health authorities get wind of the situation, and an all-out scramble begins to determine where the disease started and how it's spreading. Barbara Han, a code-writing ecologist, hopes her algorithms will put an end to that reactive model. She wants to predict outbreaks and enable authorities to prevent the next pandemic. Han takes a big-data approach, using a machine-learning AI to identify the wild animal species that carry zoonotic diseases and transmit them to humans.

Google DeepMind's AI Beats Humans At Even More Computer Games 96

An anonymous reader writes: Google DeepMind's learning algorithm has trumped human performance in an even greater range of games from the Atari 2600. The system's performance in classic games for the 80's games console has improved steadily since it was revealed in April last year (video) and a paper released yesterday shows it besting people in 31 titles.

'Rose' Wins 2015 Loebner Contest, But Big Prize Remains Unclaimed 58

The Next Web reports that developer Bruce Wilcox created the most convincing bot entered in this year's annual Loebner Competition. His latest entry, a chatbot named Rose, passed itself (herself?) off as a 30-year-old security consultant well enough to fool judges for a few minutes. But Wilcox's first-place entry was still not good enough to win the $100,000 Loebner Prize, to be given only for a more convincing impersonation. The article notes: "This isn't Wilcox’s first entry – or win. In 2010, he took first place with a bot named 'Suzette,' and followed that up in 2011 with another win using a new bot called 'Rosette.'"

Barbie Gets a Brain 235

minstrelmike writes: Mattel is coming out with a Talking Barbie designed by a huge team and pre-scripted with thousands of responses controlled by an AI, with designs to be your best friend. The design team remembers the "Math is hard" debacle of the 1990s and if a girl asks if she's pretty, Barbie will respond, "Yes. And you're smart, too." If she asks if Barbie believes in God, she says a person's beliefs are personal. And suggests talking to grownups about some problems. The linked New York Times' article ("Barbie Wants to Get to Know Your Child") even discusses trying to avoid edited vids on YouTube by scripting out words such as "cockroach."

Microsoft's Satya Nadella Shown Up By Confused Cortana Assistant 201

An anonymous reader writes: Microsoft CEO Satya Nadella was a little embarrassed at a Salesforce conference today when he tested the company's personal virtual assistant during a presentation. Slightly fluffing the question 'Show me my most at-risk opportunities', Nadella was dismayed to find Cortana offering him a Bing page with the search term 'Show me to buy milk at this opportunity'. Two further efforts to discover the exposure of his shares failed to achieve their aim, and eventually the CEO of Microsoft gave up. The fact that he stumbled over his first attempt at the question seemed to floor Cortana, which uses the 'Einstein' AI engine, and which has been more praised for its accurate speech recognition than its ability to understand what an array of interpreted words actually mean.

Researcher Trying To Teach Computer What Women He's Attracted To 181

jfruh writes: Harm de Vries, a post-doctoral researcher at the Université de Montréal, is trying to build an algorithm that will sort through pictures on Tinder and OKCupid and pick out women he'll find attractive. "Tinder kept giving me pictures of girls I wasn't attracted to," he said in a phone interview. "So I wondered if I could use deep learning." His program, built using deep learning techniques, has about a 68 percent success rate, which isn't that bad. (A human friend to whom de Vries described his preferences managed 76 percent.)

Neural Network Chess Computer Abandons Brute Force For "Human" Approach 95

An anonymous reader writes: A new chess AI utilizes a neural network to approach the millions of possible moves in the game without just throwing compute cycles at the problem the way that most chess engines have done since Von Neumann. 'Giraffe' returns to the practical problems which defeated chess researchers who tried to create less 'systematic' opponents in the mid-1990s, and came up against the (still present) issues of latency and branch resolution in search. Invented by an MSc student at Imperial College London, Giraffe taught itself chess and reached FIDE International Master level on a modern mainstream PC within three days.

The First Talking, Artificially Intelligent Surveillance Camera 58

merbs writes: Two NYU AI researchers have created a surveillance camera that, when hooked up to a crude artificial intelligence, speaks aloud what it 'sees'. "Our idea was to raise awareness regarding the omnipresence of surveillance equipment, and the current state of technological advancement with artificial intelligence," Ross Goodwin said. "We wanted to create an entity with its own sense of social awareness, its own eyes, and an ability to communicate with humans, albeit with some glitchiness that underscores the limitations of the current technology."

Software Takes On School Science Tests In Search For Common Sense 44

holy_calamity writes: Making software take school tests designed for human kids can help the quest for machines with common sense, says researchers at the Allen Institute for Artificial Intelligence. They've made software called Aristo that scores 75 percent on the multiple choice questions that make up most of New York State's 4th grade science exam. The researchers are urging other researchers to pit their best software against school tests, too, to provide a way to benchmark progress and spur competition.

Facebook Thinks Occlusion Is the Next Great Frontier For Image Recognition 32

An anonymous reader writes: Researchers at Facebook AI Research (FAIR) have published a paper contending that image recognition research is now advanced enough to consider the problem of occlusion, wherein the objects AI must identify are either partially cropped or partially hidden. Their solution is the predictably labor-expensive route of human annotation of existing image-set databases, in this case 'finishing off' occluded objects with vector outlines and assigning them a z-order. This article looks at the practical and even philosophical problems of getting IR algorithms to 'guess' objects usefully, and asks whether practical IR research might not be currently limited both by the use of over-specific image datasets and — in the field of neural networks — by problems of theory and limited 'local' processing power in critical real-time situations.

Ashley Madison Source Code Shows Evidence They Created Bots To Message Men 311

An anonymous reader writes: Gizmodo's Annalee Newitz looked through the source code contained in the recent Ashley Madison data dump and found evidence that the company created tens of thousands of bot accounts designed to spur their male users into action by sending them messages. "The code tells the story of a company trying to weave the illusion that women on the site were plentiful and eager." The evidence suggests bots sent over 20 million messages on the website, and chatted with people over 11 million times. The vast majority of fake accounts — 70,529 to 43 — pretended to be female, and the users targeted were almost entirely men. Comments left in the code indicate some of the issues Ashley Madison's engineers had to solve: "randomizing start time so engagers don't all pop up at the same time" and "for every single state that has guest males, we want to have a chat engager." The AI was unsophisticated, though one type of bot would try to convince men to pay and then pass them to a real person.

How Artificial Intelligence Can Fight Air Pollution In China 50

An anonymous reader writes: IBM is testing a new way to help fix Beijing's air pollution problem with artificial intelligence. Like many other cities across the country, the capital is surrounded by many coal burning factories. However, the air quality on a day-to-day basis can vary because of a number of reasons like industrial activity, traffic congestion, and the weather. IBM is testing a computer system capable of learning to predict the severity of air pollution several days in advance using large quantities of data from several different models. "We have built a prototype system which is able to generate high-resolution air quality forecasts, 72 hours ahead of time," says Xiaowei Shen, director of IBM Research China. "Our researchers are currently expanding the capability of the system to provide medium- and long-term (up to 10 days ahead) as well as pollutant source tracking, 'what-if' scenario analysis, and decision support on emission reduction actions."

Deep Learning Pioneer On the Next Generation of Hardware For Neural Networks 45

An anonymous reader writes: While many recognize Yann LeCun as the father of convolutional neural networks, the momentum of which has ignited artificial intelligence at companies like Google, Facebook and beyond, LeCun has not been strictly rooted in algorithms. Like others who have developed completely new approaches to computing, he has an extensive background in hardware, specifically chip design and this recognition of specialization of hardware, movement of data around complex problems, and ultimately core performance, has proven handy. He talks in depth this week about why FPGAs are coming onto the scene as companies like Google and Facebook seek a move away from "proprietary hardware" and look to "programmable devices" to do things like, oh, say, pick out a single face of one's choosing from an 800,000 strong population in under five seconds.

Facebook Is Now Working On Its Own Digital Assistant Called M 56

Mark Wilson writes: Sounding like a character from a James Bond movie, M is Facebook's personal digital assistant. Ready to compete with the likes of Cortana, M will live inside Facebook Messenger and take artificial intelligence a step further. Rather than just helping you to find information or create calendar entries, M will actually perform tasks on your behalf.

Once up and running, M will be able to book restaurants for you, purchase shopping, and more. It will also be possible to use the service to ask for advice — such as looking for somewhere to visit nearby, or gift suggestions — and Facebook says the AI behind M is "trained and supervised by people".

Video More From Tim O'Reilly about the 'WTF?!' Economy (Videos) 61

More From Tim O'Reilly about the 'WTF?!' Economy (Video) On August 12 we ran two videos of Tim O'Reilly talking with Slashdot's Tim Lord about changes in how we work, what jobs we do, and who profits from advances in labor-saving technology. Tim (O'Reilly, that is) had written an article titled, The WTF Economy, which contained this paragraph:

"What do on-demand services, AI, and the $15 minimum wage movement have in common? They are telling us, loud and clear, that we’re in for massive changes in work, business, and the economy."

We're seeing a shift from cabs to Uber, but what about the big shift when human drivers get replaced by artificial intelligence? Ditto airplane pilots, burger flippers, and some physicians. WTF? Exactly. Once again we have a main video and a second one available only in Flash (sorry about that), along with a text transcript that covers both videos. Good thought-provoking material, even if you think you're so special that no machine could possibly replace you.

IBM 'TrueNorth' Neuro-Synaptic Chip Promises Huge Changes -- Eventually 97

JakartaDean writes: Each of IBM's "TrueNorth" chips contains 5.4 billion transistors and runs on 70 milliwatts. The chips are designed to behave like neurons—the basic building blocks of biological brains. Dharmenda Modha, the head of IBM's cognitive computing group, says a system of 24 connected chips simulates 48 million neurons, roughly the same number rodents have.

Whereas conventional chips are wired to execute particular "instructions," the TrueNorth juggles "spikes," much simpler pieces of information analogous to the pulses of electricity in the brain. Spikes, for instance, can show the changes in someone's voice as they speak—or changes in color from pixel to pixel in a photo. "You can think of it as a one-bit message sent from one neuron to another." says one of the chip's chief designers. The chips are designed well not for training neural networks, but for executing them. This has significant implications for consumer AI: big companies with lots of resources could focus on the training, which individual TrueNorth chips in people's gadgets could handle the execution.

Google Research Leads To Automated Real-Time Pedestrian Detection 57

An anonymous reader writes with a link to a story about one of the unexciting but vital bits of technology that will need to be even further developed as autonomous cars' presence grows: making sure that those cars don't hit people. Google researchers have recently presented findings about a method that tops previous ones for real-time pedestrian detection using neural nets "that is both extremely fast and extremely accurate." From the article: There are other approaches that provide a real-time solution on the GPU but in doing so, have not achieved accuracy targets (in this real-time approach there was a miss rate of 42% on the Caltech pedestrian detection benchmark). Another approach called the VeryFast method can run at 100 frames per second (compared to the Google team's 15) but the miss rate is even greater. Others that emphasize accuracy, even with GPU acceleration, are up to 195 times slower.