Findo wanted to use the results achieved in image analysis with deep statistical models and to apply them to text analysis. The problem? Text data is extremely sparse: the more discrete the data, the more data is required to successfully train statistical models.
Along with the VAE model, Findo is researching Ladder networks as well for semi-supervised approach. These have been shown to have great potential with the semi-supervised approach within the fields, where you can have access to small amount of labeled data and large amount of unlabeled data.
Variational Autoencoders are one of the more interesting generative models Findo is working on. Basically, they combine ideas from deep learning with statistical inference. They can be used to learn a low dimensional representation (z) of high dimensional data (x) like images. X and Z refer to random variables here, unlike in standard auto encoders.
At the World Economic Forum in Davos this year, the topic of note was Big Data. The digital revolution is quickly being heralded as the fourth industrial revolution, with the physical, digital and biological worlds merging through technology. Indeed, areas like robotics, artificial intelligence and the Internet of Things have seen some overwhelming progress.
Artificial intelligence is defined as technology displaying intelligence that is normally the domain of humans. Some estimate that computers will have outpaced human intelligence by 2030. Will machines take over our jobs?
Many assistant programs are now being sold in the market such as Amazon’s Alexa, Apple’s Siri and Sound Hound, that have come about to make our lives better and easier, and a few other startups that are still trying to make their way into the market.