The research, carried out by Aditya Khosla of MIT’s CSAIL with friends from Ebay and DigitalGlobe, uses computer vision and machine learning to answer the age old question “What makes an image popular?” The researchers began with a massive corpus of 2.3 million images from Flickr, and then used deep learning (neural networks) and analysis of social cues to discover what exact combination of image features and social context makes an image likable. [Research paper, PDF]
First the software looks at high-level features, namely objects. Modern computer vision algorithms are very good at picking out and identifying objects in images. Unsurprisingly, objects such as miniskirts, bras, and guns have a strong positive impact on an image’s popularity. Most cuddly animals score a “medium positive impact” rating, while wild boars, dull-looking items of food, and other everyday objects create a “low positive impact.” Spatulas, plungers, laptops, and golfcarts actually create a negative impact.
This is especially important, given how we’re pumping so many greenhouse gases into the environment.
by Scott K. Johnson - Jan 11 2014, 8:10am PSTThe Putorana Plateau, capped basalt released in the Siberian Traps eruptions. Wikimedia Commons
Climate models usually end up in the news because of projections of future climate, but many researchers use the models to study other planets or the Earth’s past. They can help test hypotheses about past climate events by comparing model simulations to estimates of past climates obtained from things like ice and sediment cores.
One climatic event that looms large in Earth’s history is the end-Permian mass extinction about 252 million years ago—the worst mass extinction event on record. A volcanic event seems to have been at least partly to blame. Tremendously vast eruptions in Siberia coughed up lava flows and ash that may have covered an area nearly as large as Australia—a feature known as the Siberian Traps. During this event, some 90 percent of marine species disappeared, and species on land didn’t fare well, either.
Could have put this under “Wingnuts,” but this happened awhile ago, so I thought “History” was more fitting. Anyone on the right who wants to claim Republicans or conservatives never supported the Apartheid regime, here’s something else you can use to debunk them. This was also motivated by profit, so whoever says greed is good, has been proven wrong in this case as well. That also argues against these people simply being Wingnuts. It sounds like the folks at ALEC were motivated more by the fear of the lose of profits than ideology.
As the movement for public and private divestment from apartheid South Africa grew throughout the United States in the 1980s, the American Legislative Exchange Council (ALEC) aggressively mobilized against South African divestment, stymying state and federal efforts to sanction, isolate, and divest from the Pretorian regime, according to documents newly uncovered by People For the American Way and the Center For Media and Democracy.
ALEC used state and federal policy papers, monthly newsletters, “fact-finding” missions, panel discussions led by lobbyists on the payroll of the South African apartheid regime, and other means to pursue an anti-divestment agenda, one that relied solely on “corporate beneficence” to pressure the country to reform. This effort, in turn, was funded by corporations that were heavily invested in South Africa and had the most to lose from divestment.
In other words, if I make four calls from four different places over the course of a 15-month period, my pattern of movement could be identified out of a population two and a half times the size of Washington, D.C.
The man who invented Roomba, the robotic vacuum, is back — this time, with Baxter. Rodney Brooks, roboticist and entrepreneur, brought Baxter, his latest robot, to the TED conference in Long Beach, Calif., last week.
Brooks’ latest company, Rethink Robotics, describes Baxter as a collaborative manufacturing robot. Brooks showed how Baxter, which costs $22,000 per model, can work alongside humans — not replace them — to do simple, repetitive tasks.
Brooks is the Panasonic Professor of Robotics (emeritus) at MIT, where he also used to run the Computer Science and Artificial Intelligence Laboratory. His idea is that Baxter can help, for example, aging factory workers do their jobs more efficiently. He told the TED audience that after this generation ages out of factory work, they tell him, they don’t want their children to carry on their work.
Baxter has eyes for feedback. Though its eyes don’t see you as a person, they serve as a signal as to what it’ll do next.
Normally robots need to be programmed, but this one learns by physical training. Move Baxter’s arm and it learns that’s how it should move its arm. In just a few minutes, the robot can be taught, for example, to take something out of a box and place it on a conveyor belt, then, after it’s assembled, put it back in a box.
Massachusetts Institute of Technology security spotted an armed man on its campus. So far there has been not shooting reported. Lets hope that there is no violence and this is settled without any injury or killing. From yahoo: news.yahoo.com
Update from CNN and local police: “Scene is clear. Call unfounded. No threat to public safety…” news.blogs.cnn.com
Earthworms creep along the ground by alternately squeezing and stretching muscles along the length of their bodies, inching forward with each wave of contractions. Snails and sea cucumbers also use this mechanism, called peristalsis, to get around, and our own gastrointestinal tracts operate by a similar action, squeezing muscles along the esophagus to push food to the stomach.
Now researchers at MIT, Harvard University and Seoul National University have engineered a soft autonomous robot that moves via peristalsis, crawling across surfaces by contracting segments of its body, much like an earthworm. The robot, made almost entirely of soft materials, is remarkably resilient: Even when stepped upon or bludgeoned with a hammer, the robot is able to inch away, unscathed.
Sangbae Kim, the Esther and Harold E. Edgerton Assistant Professor of Mechanical Engineering at MIT, says such a soft robot may be useful for navigating rough terrain or squeezing through tight spaces.
We used to think about machines taking over mundane jobs, like twisting a screw into a toaster on an assembly line over and over again. But more recently, technology is eliminating higher-skill jobs.
To talk about this, some of the nation’s top technologists and economists came to Harvard and the Massachusetts Institute of Technology this week for a jobs conference called Race Against the Machine. There was a very impressive machine on hand: Watson, the powerful Jeopardy-playing computer built by IBM that recently beat the all-time human Jeopardy game show champion.
This week Watson took on two teams from MIT and Harvard. The auditorium was packed with students. Although Harvard managed to do well, its team was no match for the powerful computer.
Only tepid applause greeted Watson after it correctly answered questions about things like a truncated icosahedron (soccer ball-shaped). Maybe the reason human beings don’t tend to clap for Watson is that we can instinctively sense that it might steal our jobs.