The efforts of a team of scientists, hired by the famous Google X Labs, have been successful in the direction of crating the first artificial brain, capable of learning from its own experience.
Rather famous for inventing the self-driving cars and for the Google Project Glass (augmented reality glasses), the scientists of Google X Labs are currently working on an new, highly-ambitious project, that attempts to create a virtual replica of the human brain, capable of learning.
In order to reach their goal, the Google researchers have created an artificial neural network, based on a processing machine that has 16,000 interconnected processors and underpinned by a specialized experimental software. The prototype that resulted was then connected to the network and let run free on the internet, while the researchers team monitored the results and its behavior.
Among others, during the experiment the neural network learned how to identify cat pictures, accurately extracted from a collection of over 10 billion miniatures created using Youtube videos. The results seem to support the biologists’ theories that suggest that the brain’s neurons have a natural ability to identify specific objects.
What is really remarkable is that the scientists haven’t introduced any information into the neural network, without suggesting it what to find. The system independently invented the “concept of cat,” without any information coming from outside.
In the first instance, the machines capable of learning can be used in the search engine field, improving the relevance and accuracy of the results provided for the semantic searches – when the users are basically asking questions in the search engine, receiving relevant responses for the context, instead of words used as web search query.
Improvements can also be brought to the translation algorithms, voice recognition systems, translating conversation from one language to another and face recognition systems. View the original article here
No comments:
Post a Comment