Tag: Graph Search

Looking for Time Travelers… on Twitter!

DoctorWho_TardisIf time travelers were real, a la Doctor Who or Doc Emmett Brown style, how would they go about sharing their gift with the world? According to astrophysicist Robert Nemiroff and physics graduate student Teresa Wilson at Michigan Technological University, they would tweet about it. And so, the two began what has proven to be one of the most interesting searches on today’s social media.

To break it down succinctly, the pair began to search the backlogs of Facebook and Twitter for any indications of time travellers posting about the future. This they did by entering search terms for two major events – the appearance Comet ISON in September of 2012 and the election of Pope Francis in March 2013 – to see if there was any mention of them before they happened.

comet_ISONTheir theory, as presented in a paper published last month on Cornell University’s Library website, was that if there were any postings containing “Comet ISON,” “#cometison,” “Pope Francis” or “#popefrancis” from before those dates, they may very well be from a time traveler. Unfortunately, their searchers on Facebook turned up results which, in their own words, “were clearly not comprehensive.

Granted, a time-traveler would be quick to delete any status updates that appeared prescient, using Facebook’s new Graph Search privacy features. However, the the time-traveler hunting due had no better luck on Twitter, where a majority of people keep their tweets public. But of course, they went on to say in their paper that just because they didn’t see any time travelers doesn’t mean they don’t exist.

time-slipAs they argue it, it might not be possible for time travelers to leave any evidence of their journey behind.

…it may be physically impossible for us to find such information as that would violate some yet-unknown law of physics… time travelers may not want to be found, and may be good at covering their tracks.

Another thing to consider is that time travelers might actively try to erase any mention of their existence. For example, in the first season of the Doctor Who reboot, where the Doctor used a special virus to delete any digital trace of himself before leaving the present age. It’s academic stuff really, and the pair really shouldn’t have expected such careless errors to show up on the internet.

the_time_machine_croppedAnd as Donald Rumsfeld, another man who went searching for something and came up empty, said: “the absence of evidence is not the evidence of absence”. Yes, I know, the comparison really doesn’t help, does it? And right now, I’m sure you might be wondering if all those tax dollars that fund research grants might be better used elsewhere.

And let’s face it, it’s something many of us would wonder, and possible check for ourselves, given half a chance…

Sources: universetoday.com, huffingtonpost.com

The Future of Computing: Brain-Like Computers

neuronsIt’s no secret that computer scientists and engineers are looking to the human brain as means of achieving the next great leap in computer evolution. Already, machines are being developed that rely on machine blood, can continue working despite being damaged, and recognize images and speech. And soon, a computer chip that is capable of learning from its mistakes will also be available.

The new computing approach, already in use by some large technology companies, is based on the biological nervous system – specifically on how neurons react to stimuli and connect with other neurons to interpret information. It allows computers to absorb new information while carrying out a task, and adjust what they do based on the changing signals.

brain_chip2The first commercial version of the new kind of computer chip is scheduled to be released in 2014, and was the result of a collaborative effort between I.B.M. and Qualcomm, as well as a Stanford research team. This “neuromorphic processor” can not only automate tasks that once required painstaking programming, but can also sidestep and even tolerate errors, potentially making the term “computer crash” obsolete.

In coming years, the approach will make possible a new generation of artificial intelligence systems that will perform some functions that humans do with ease: see, speak, listen, navigate, manipulate and control. That can hold enormous consequences for tasks like facial and speech recognition, navigation and planning, which are still in elementary stages and rely heavily on human programming.

googleneuralnetworkFor example, computer vision systems only “recognize” objects that can be identified by the statistics-oriented algorithms programmed into them. An algorithm is like a recipe, a set of step-by-step instructions to perform a calculation. But last year, Google researchers were able to get a machine-learning algorithm, known as a “Google Neural Network”, to perform an identification task (involving cats) without supervision.

And this past June, the company said it had used those neural network techniques to develop a new search service to help customers find specific photos more accurately. And this past November, researchers at Standford University came up with a new algorithm that could give computers the power to more reliably interpret language. It’s known as the Neural Analysis of Sentiment (NaSent).

deep_learning_laptopA similar concept known as Deep Leaning is also looking to endow software with a measure of common sense. Google is using this technique with their voice recognition technology to aid in performing searches. In addition, the social media giant Facebook is looking to use deep learning to help them improve Graph Search, an engine that allows users to search activity on their network.

Until now, the design of computers was dictated by ideas originated by the mathematician John von Neumann about 65 years ago. Microprocessors perform operations at lightning speed, following instructions programmed using long strings of binary code (0s and 1s). The information is stored separately in what is known as memory, either in the processor itself, in adjacent storage chips or in higher capacity magnetic disk drives.

neural-networksBy contrast, the new processors consist of electronic components that can be connected by wires that mimic biological synapses. Because they are based on large groups of neuron-like elements, they are known as neuromorphic processors, a term credited to the California Institute of Technology physicist Carver Mead, who pioneered the concept in the late 1980s.

These processors are not “programmed”, in the conventional sense. Instead, the connections between the circuits are “weighted” according to correlations in data that the processor has already “learned.” Those weights are then altered as data flows in to the chip, causing them to change their values and to “spike.” This, in turn, strengthens some connections and weakens others, reacting much the same way the human brain does.

Neuromorphic-chip-640x353In the words of Dharmendra Modha, an I.B.M. computer scientist who leads the company’s cognitive computing research effort:

Instead of bringing data to computation as we do today, we can now bring computation to data. Sensors become the computer, and it opens up a new way to use computer chips that can be everywhere.

One great advantage of the new approach is its ability to tolerate glitches, whereas traditional computers are cannot work around the failure of even a single transistor. With the biological designs, the algorithms are ever changing, allowing the system to continuously adapt and work around failures to complete tasks. Another benefit is energy efficiency, another inspiration drawn from the human brain.

IBM_stacked3dchipsThe new computers, which are still based on silicon chips, will not replace today’s computers, but augment them; at least for the foreseeable future. Many computer designers see them as coprocessors, meaning they can work in tandem with other circuits that can be embedded in smartphones and the centralized computers that run computing clouds.

However, the new approach is still limited, thanks to the fact that scientists still do not fully understand how the human brain functions. As Kwabena Boahen, a computer scientist who leads Stanford’s Brains in Silicon research program, put it:

We have no clue. I’m an engineer, and I build things. There are these highfalutin theories, but give me one that will let me build something.

calit2PhotoLuckily, there are efforts underway that are designed to remedy this, with the specific intention of directing that knowledge towards the creation of better computers and AIs. One such effort comes from the National Science Foundation financed the Center for Brains, Minds and Machines, a new research center based at the Massachusetts Institute of Technology, with Harvard and Cornell.

Another is the California Institute for Telecommunications and Information Technology (aka. Calit2) – a center dedicated to innovation in nanotechnology, life sciences, information technology, and telecommunications. As
Larry Smarr, an astrophysicist and director of Institute, put it:

We’re moving from engineering computing systems to something that has many of the characteristics of biological computing.

Human-Brain-project-Alp-ICTAnd last, but certainly not least, is the Human Brain Project, an international group of 200 scientists from 80 different research institutions and based in Lausanne, Switzerland. Having secured the $1.6 billion they need to fund their efforts, these researchers will spend the next ten years conducting research that cuts across multiple disciplines.

This initiative, which has been compared to the Large Hadron Collider, will attempt to reconstruct the human brain piece-by-piece and gradually bring these cognitive components into an overarching supercomputer. The expected result of this research will be new platforms for “neuromorphic computing” and “neurorobotics,” allowing for the creation of computing and robotic architectures that mimic the functions of the human brain.

neuromorphic_revolutionWhen future generations look back on this decade, no doubt they will refer to it as the birth of the neuromophic computing revolution. Or maybe just Neuromorphic Revolution for short, but that sort of depends on the outcome. With so many technological revolutions well underway, it is difficult to imagine how the future will look back and characterize this time.

Perhaps, as Charles Stross suggest, it will simply be known as “the teens”, that time in pre-Singularity history where it was all starting to come together, but was yet to explode and violently change everything we know. I for one am looking forward to being around to witness it all!

Sources: nytimes.com, technologyreview.com, calit2.net, humanbrainproject.eu

Judgement Day Update: Deep Learning and AIs

brain_chip2Moore’s Law, a sort of general law of computing that was coined by Intel co-founder Gordon E. Moore, states that the number of transistors on integrated circuits doubles approximately every two years. In accordance with this observation, which was made back in 1965, computers have evolved to the point where they are over 650 times faster than they were in the early 1970’s.

Given this trend, scientists and futurists have long suspected that it is only a matter of time before computers surpass human beings in terms of raw intelligence. Already, supercomputers like IBM’s Watson, University of Illinois’ “Blue Waters”, and MIT’s ConceptNet 4 have demonstrated that they are superior in terms of raw knowledge and retention. But when it comes to actual thinking – i.e. independent reasoning and common sense – they are seriously lacking.

IBM_Blue_Gene_P_supercomputerYes, despite all the progress made in the field of computing in the past 40 years, bridging that gap between computational power and artificial intelligence has remained elusive. And at this point, no one is quite sure what it will take to make that leap happen. However, a number of developers and designers are coming up with breakthroughs that may constitute a small jump.

One such breakthrough comes from Standford University, where computer engineers have come up with a new algorithm that could give computers the power to more reliably interpret language. Called Neural Analysis of Sentiment – or NaSent for short – the algorithm seeks to improve on current methods of written language analysis by drawing inspiration from the human brain.

AI_picNaSent is part of a movement in computer science known as deep learning, a new field that seeks to build programs that can process data in much the same way the brain does. According to Richard Socher, the Stanford University grad student who helped develop NaSent:

In the past, sentiment analysis has largely focused on models that ignore word order or rely on human experts. While this works for really simple examples, it will never reach human-level understanding because word meaning changes in context and even experts cannot accurately define all the subtleties of how sentiment works. Our deep learning model solves both problems.

Socher was joined by artificial-intelligence researchers Chris Manning, a Professor of Computer Science and Linguistics at Standford, and Andrew Ng, one of the engineers behind Google’s deep learning project and who is working on creating the “Google Brain”.  Their aim is to develop algorithms that can operate without continued help from humans.

word_cloudWhat sets this algorithm apart is its ability to identify the meaning of words in the context of phrases and sentences. This is a big change from previous methods of sentiment analysis which have been limited to parsing through a collection of words and ranking them as either positive or negative without taking word order into account.

To create NaSent, Socher and his team used 12,000 sentences taken from the movie reviews website Rotten Tomatoes. They split these sentences into roughly 214,000 phrases that were labeled as very negative, negative, neutral, positive, or very positive, and then they fed this labeled data into the system, which NaSent then used to predict whether sentences were positive, neutral or negative on its own.

metadataAccording to the researchers, NaSent is about 85 percent accurate, a 5% improvement over previous models. However, Socher and his team are working on increasing that rate of success by feeding the system more data from Twitter and the Internet Movie Database, and have also set up a live demo where people can type in their own sentences.

And whereas the field of deep learning began in the academic world, it has since spread to web giants such as Google and Facebook. In both cases, the companies have taken to hiring people familiar with the field to help them sort through the growing mountains of data they are indexing to help them improve their products, particularly those that already rely on machine learning.

In Google’s case, these include voice recognition technology, as well as their ongoing work in neural networks. In the case of Facebook, they are looking to deep learning to help them improve Graph Search, a search engine that allows people to search activity on their network. This presents a challenge, mainly because machines have a hard time interpreting the nuances of human language. Just ask Watson, IBM’s own supercomputer!

Sources: wired, (2), (3), engineering.stanford.edu