Saturday, May 22, 2010

May 21 - Darpa’s Self-Learning Software Knows Who You Are‏

Dear Friends,

paste/click the link if you can't access the links/video.

Be Well.


Darpa’s Self-Learning Software Knows Who You Are

Software systems could one day analyze everything from blurry war-zone footage to the subtle sarcasm in a written paragraph, thanks to two unassuming scientists who are inspired by biology to make revolutionary strides in intelligent computing.

Yann LeCun and Rob Fergus, both computer science professors at New York University, are the brains behind “Deep Learning,” a program sponsored by Darpa, the Pentagon’s blue-sky research agency. The idea, ultimately, is to develop code that can teach itself to spot objects in a picture, actions in a video, or voices in a crowd. LeCun and Fergus have $2 million and four years to make it happen.

Existing software programs rely heavily on human assistance to identify objects. A user extracts key feature sets, like edge statistics (how many edges an object has, and where they are) and then feeds the data into a running algorithm, which uses the feature sets to recognize the visual input.

“People spend huge amounts of time building these feature sets, figuring out which are better or more accurate, and then refining them,” LeCun told Danger Room. “The question we’re asking is whether we can create computers that automatically learn feature sets from data. The brain can do it, so why not machines?”

The computer systems will be inspired by biology, but not modeled after it. That’s because researchers still aren’t entirely sure how animals are able to turn inputs — an object, a movement, a sound — into usable information. Ten years ago, a study at MIT helped answer the question. Researchers rewired ferret brains, so that the optical nerve fed into the auditory cortex, and vice versa. But the ferrets still saw and heard normally, leading the team to conclude that brain function depends on the signal — not the area.

Brains also display plenty of abstraction when it comes to identifying specific inputs: LeCun was inspired to create his algorithmic layering approach, called “a convolutional network,” by the 1960s research of David Hubel and Torstein Weisel. The two used cats to demonstrate how the brain’s visual cortex relies on abstractions to create complex representations of a given visual input.

In other words, LeCun said, “There’s some sort of learning algorithm within the brain. We just don’t know what it is.”


But the algorithmic talents of the mind, along with its ability to identify visual data by abstraction, will be the key components of the NYU team’s new system. Right now, an algorithm recognizes objects in one of two ways. In one, it is shown some representative examples of what, say, a horse looks like. Then the code tries to match any new creature to the ur-stallion. (That’s called “supervised” learning.) In the other way, the software is shown lots and lots of horses, and it builds its own model of what a horse is supposed to resemble. (That’s “unsupervised” learning.)

What LeCun and Fergus are trying to do is make code that can get it right on a first, unsupervised example — using layer after layer of code to abstract the essential attributes of an object. This first step is to turn an image into numbers: For a 100 x 100 pixel image, the software produces a grid of 10,000 numbers; 9 x 9 “masks” are then applied to that grid, to uncover attributes of the image. The first feature spotted is an object’s edge. (The human brain makes a similar initial pass.) Several more “masks” follow. The final output? A series of 256 numbers that identifies the input.

The two are only six weeks into the project, but they’ve already got demos up and running.

The Deep Learning algorithm and I had never met, but with a quick shot by a small webcam on LeCun’s laptop, the layers of code captured my features and could immediately distinguish me from other objects and people in LeCun’s office. The same thing happens when LeCun introduces the system to two different coffee mugs — it takes mere seconds for the computer to acquaint itself with each, then distinguish one from the other.

And this is only the beginning. Darpa also wants a system that can spot activities, like running, jumping or getting out of a car. The final version will operate unsupervised, by being programmed to hold itself accountable for errors — and then auto-correct them at each algorithmic layer.

It should also be able to apply the layered algorithmic technique to text. Right now, computer systems can parse sentences to categorize them as positive or negative, based on how often different words appear in the text. By applying layers of analysis, the Deep Learning machine will — LeCun and Fergus hope — spot sarcasm and irony too.

“Ideally, what we’ll come away with is a ‘generic learning box’ that can identify every data cue,” Fergus tells Danger Room.

Photo: Katie Drummond

No comments:

Post a Comment



Click upon the circle after the small square for captions


How to Digitally Record/Video a UFO sighting:

Como registar digitalmente ou gravar um vídeo de um avistamento de um UFO:

Stabilize the camera on a tripod. If there is no tripod, then set it on top of a stable, flat surface. If that is not possible lean against a wall to stabilize your body and prevent the camera from filming in a shaky, unsteady manner.

Estabilize a camera com um tripé. Se não tiver um tripé, então coloque-a em cima de uma superfície estável. Se não for possível, então encoste-se a uma parede para estabilizar o corpo e evitar que a camera registe de maneira tremida e instável.

Provide visual reference points for comparison. This includes the horizon, treetops, lampposts, houses, and geographical landmarks (i.e., Horsetooth Reservoir, Mt. Adams, etc.) Provide this in the video whenever is appropriate and doesn’t detract from what your focus is, the UFO.

Forneça pontos visuais de referência para comparação. Isso inclui o horizonte, cimo das árvores, postes de iluminação, pontos de referência geográficos (como o Reservatório de Horsetooth, Mone Adams, etc) Forneça esses pontos no vídeo sempre que for apropriado e não se distraia do que é o seu foco, o UFO/a Nave.

Narrate your videotape. Provide details of the date, time, location, and direction (N,S,E,W) you are looking in. Provide your observations on the weather, including approximate temperature, windspeed, any visible cloud cover or noticeable weather anomalies or events. Narrate on the shape, size, color, movements, approximate altitude of the UFO, etc and what it appears to be doing. Also include any unusual physical, psychological or emotional sensations you might have. Narrate any visual reference points on camera so they correlate with what the viewer will see, and thereby will be better able to understand.

Faça a narração do vídeo. Forneça pormenores sobre a data, hora, local e direcção (Norte, Sul, Este, Oeste) que está a observar. Faça observações sobre as condições atmosféricas, incluindo a temperatura aproximada, velocidade do vento, quantidade de nuvens, anomalias ou acontecimentos meteorológicos evidentes. Descreva a forma, o tamanho, a cor, os movimentos, a altitude aproximada onde se encontra o UFO/nave, etc e o que aparenta estar a fazer. Inclua também quaisquer aspectos pouco habituais de sensações físicas, psicológicas ou emocionais que possa ter. Faça a narração de todos os pontos de referência visual que o espectador irá ver e que, deste modo, será capaz de compreender melhor.

Be persistent and consistent. Return to the scene to videotape and record at this same location. If you have been successful once, the UFO sightings may be occurring in this region regularly, perhaps for specific reasons unknown, and you may be successful again. You may also wish to return to the same location at a different time of day (daylight hours) for better orientation and reference. Film just a minute or two under “normal” circumstances for comparison. Write down what you remember immediately after. As soon as you are done recording the experience/event, immediately write down your impressions, memories, thoughts, emotions, etc. so it is on the record in writing. If there were other witnesses, have them independently record their own impressions, thoughts, etc. Include in this exercise any drawings, sketches, or diagrams. Make sure you date and sign your documentation.

Seja persistente e não contraditório. Volte ao local da cena e registe o mesmo local. Se foi bem sucedido uma vez, pode ser que nessa região ocorram avistamentos de UFOs/naves com regularidade, talvez por razões específicas desconhecidas, e talvez possa ser novamente bem sucedido. Pode também desejar voltar ao mesmo lugar a horas diferentes do dia (durante as horas de luz)para ter uma orientação e referência melhor. Filme apenas um ,inuto ou dois em circunstâncias “normais” para ter um termo de comparação. Escreva tudo o que viu imediatamente após o acontecimento. Logo após ter feito o registo da experiência/acontecimento, escreva imediatamente as impressões, memórias, pensamentos, emoções, etc para que fiquem registadas por escrito. Se houver outras testemunhas, peça-lhes para registar independentemente as suas próprias impressões, pensamentos, etc. Inclua quaisquer desenhos, esbolos, diagramas. Certifique-se que data e assina o seu documento/testemunho.

Always be prepared. Have a digital camera or better yet a video camera with you, charged and ready to go, at all times. Make sure you know how to use your camera (and your cell phone video/photo camera) quickly and properly. These events can occur suddenly, unexpectedly, and often quite randomly, so you will need to be prepared.

Esteja sempre preparado, Tenha sempre uma camera digital, melhor ainda, uma camera vídeo consigo, carregada e pronta a usar sempre que necessário. Certifique-se que sabe como lidar com a sua camera (ou com o seu celular/camera fotográfica) rápida e adequadamente. Esses acontecimentos podem acontecer súbita e inesperadamente e, por vezes, acidentalmente, por isso, necessita estar preparado.

Look up. Be prepared. Report. Share.

Olhe para cima, Esteja preparado, Relate, Partilhe.



Pf., clique no símbolo do YouTube e depois no quadrado pequeno, em baixo, ao lado direito para obter as legendas CC, e escolha PORTUGUÊS

埋め込み画像 4埋め込み画像 5

What time is Around the World?


AND YOU AND I - click image



NGC - UFO's in EUROPE (Porugal included)

FEBRUARY 7, 2013 - 7:00PM EST

FEBRUARY 7, 2013 - 7:00PM EST