It is Black Friday tomorrow – the day after Thanksgiving in the US, and the day which traditionally marks the start of Christmas Shopping. Have you been feeling overwhelmed by the huge amount of tinsel-wrapped holiday offers and bargains? If you cannot face the shops and malls with their piped Christmas carols – and it isn’t even December yet – is it harder and harder to find good deals among the thousands of items in the online catalogs? Wouldn't it be easier if someone could just highlight great opportunities for you? Worry no more! Statistics are here to help! I know, it sounds like a fairy tale or something Santa has dreamed up to comfort you, but it is actually happening unsecured loans.
Once upon a time, in Columbia University, the Laboratory for Intelligent Imaging and Neural Computing announced an agreement with a company called Neuromatters LLC to develop a novel brain-computer interface (BCI) technology for rapid identification of relevant images. “The human brain reacts to images of interest at a pace that is far faster than a person can consciously register” – or so it is explained in the University's press release. -“ Researchers at Columbia University have developed a technology, 'Cortically Coupled Computer Vision (C3Vision)', that takes advantage of this near-subconscious ability and pairs it with the processing power and efficiency of computers for rapid identification of images that the brain finds relevant”.
In simpler terms, the developed system will constantly scan a human's brain to detect what type of images the brain’s owner finds interesting, based on the brain's neuronal activity while seeing the items. With this information, the system will “learn” what images are more important to the user, so it can automatically detect when you spot something novel or interesting, even if you’re unaware of it. Paul Sajda, PhD, Director of the Laboratory for Intelligent Imaging and Neural Computing and at Columbia University’s School of Engineering briefs it like this: “The system latches on to individual perceptions and trains the computer to know what the user means by interesting. The computer and the brain operate synergistically.”
The real beauty in the design lies in the “learning” part of the project. The process of machine learning is one of the most interesting parts of artificial intelligence nowadays, at least to statisticians. For those unaware of it, machine learning includes several statistical models developed to teach a computer to improve its performance by adjusting the parameters in the model in a way that optimize results. In particular, C3Vision uses a form of Neural Network, a non-linear statistical model specifically designed to find patterns of data. Artificial Neural Networks are used in many different fields that include military applications, commercial systems developed to help police solve crimes and even credit fraud detection... and now, brain scanning. As you may have guessed, neural network models are based on the biological structure of the brain and are part of other artificial intelligence schemes that have been inspired by nature (genetic algorithms are also a good example).
Right now, the U.S. Army is interested in C3Vision as a way to help soldiers navigate through dangerous terrains. A soldier could detect suspicious items or hazards in the road faster with this gear. The system may also extend to video surveillance and security, where the aim is to identify suspicious activity. But who cares about security? We want to shop! And it is a relief that Paul Sajda knows that very well, as he envisions the technology eventually improving civilian shopping lives as well: “There’s nothing out there that can't really use your subjective preference as a signature to guide you”, he claims. Think of a miniaturized, wireless version of the device used to tag consumers’ reactions to items or even shops as you walk down a city street. The model will then detect great shopping deals or products specifically designed for you while you walk through the mall and even offer nearby stores selling similar items suited to your interests. Tired of seeing that many wonderful best-of-the-festive-season offers? Don't worry, C3Vision will detect the interesting ads for you, so you can ignore the rest.
Donna See, the officer at Columbia Technology Ventures who manages the C3Vision patent portfolio, goes beyond that, exploring further possibilities for the system: “We can certainly envision C3Vision applied to image-rich online contexts, such as fashion, furnishings, even real estate and travel, learning the preferences of tastemakers and buyers to create truly customized and targeted retail experiences.”
So, as you can see, statistical models will achieve their ultimate goal: after predicting the economy, analyze pandemics and describe social phenomenons, a model will finally be able to help you on Black Friday. Happy Christmas shopping!
No comments:
Post a Comment