Right now, artificial intelligence is used to make snap trading decisions in response to news reports. It could write this press release. Companies are rushing to integrate AI into products and services. In the rush, it is likely many companies – especially entrepreneurial companies - will make costly mistakes. According to Tal, "As executives become enamored and anxious about adopting AI technology, there is an increased chance for making a technology bet that can consume the assets of a company. It is critical that companies map out their opportunities, test and prove a tech integration before making a huge investment that could actually derail a company strategy."

Apple is getting serious about its push towards augmented reality. The firm has filed a a patent for a mysterious iPhone'window' that could be used to view virtual objects in the real world. Currently, augmented reality features rely on the iPhone camera, but this new window might make those objects more realistic. The'hybrid' display could have a clear segment at the top, allowing users to, for instance, see virtual Pokemon, as they walk down the street. The'hybrid' display could have a clear segment at the top through which users could view virtual objects in the real world, according to a patent filing (pictured) The strange window could create transparent window regions in front and behind the device.

Newsroom in online ecosystem is difficult to untangle. With prevalence of social media, interactions between journalists and individuals become visible, but lack of understanding to inner processing of information feedback loop in public sphere leave most journalists baffled. Can we provide an organized view to characterize journalist behaviors on individual level to know better of the ecosystem? To this end, I propose Poisson Factorization Machine (PFM), a Bayesian analogue to matrix factorization that assumes Poisson distribution for generative process. The model generalizes recent studies on Poisson Matrix Factorization to account temporal interaction which involves tensor-like structure, and label information. Two inference procedures are designed, one based on batch variational EM and another stochastic variational inference scheme that efficiently scales with data size. An important novelty in this note is that I show how to stack layers of PFM to introduce a deep architecture. This work discusses some potential results applying the model and explains how such latent factors may be useful for analyzing latent behaviors for data exploration.

I have read a lot about neural networks (and it's application to vision and language) and statistical learning, now I realised there is area caled "meta-learning". Is there any site with papers that could introduce me to this field or anything that will help me to productively approach this field?

To hear Alphabet tell it, an artificial–intelligence-first future will one day surround us.

After all, the whole point of artificial intelligence is to make our gelatinous brains unnecessary. Those brains cost money to maintain, and that's not helpful for revenue multiples. But even mundane tasks, such as booking a hotel room, are difficult problems to accomplish with algorithms.

Artificial Intelligence is the fourth industrial revolution after the invention of steam power, electricity, electronics and software. And Bottos will be the key driving force for this revolution as the data becomes the core competitive power in AI development.

A galaxy far, far away is a little closer with the invention of a robotic arm inspired by Luke Skywalker's bionic hand. And while this arm may not wield a lightsaber, it has a greater power for jazz musician Jason Barnes -- it lets him play the piano for the first time in five years. Barnes, who lost much of his right arm in a work accident, is back at the keys with an AI prosthesis created by researchers at the Georgia Institute of Technology. Unlike most prosthetics, it gives the 28-year-old the ability to control each finger individually. With it, Barnes can play Beethoven.

Amazon will hold its second annual Digital Day sale on Friday Dec. 29, in which it will offer deals on games, apps, movies and music. The Digital Day sale will follow Amazon's successful Black Friday and Cyber Monday, as well as the company's record-breaking holiday season. Digital Day will offer more than 5,000 deals across the shopping platform. This year around, the company is holding 40 percent more deals than in 2016, Amazon said. The online retailer will offer up to 80 percent off digital content, which includes apps, games, TV shows, movies, digital comics and more.

A few months ago I stumbled onto an interesting idea while listening to the TWiML & AI podcast. It described a process by which one could attempt to introduce confusion into a network (starting at any arbitrary hidden layer) so that it couldn't learn from select biases in the training data. For example, if you were training an image classification network, and you wanted to forbid the network to learn anything about race, you could use this technique, to do so. The problem is that I can't for the life of me remember what this technique is called, or what episode of the podcast it was discussed in. All that I remember is that I believe that the method proposed involved choosing a layer beyond which you didn't want the network to be able to include information about the bias you were trying to remove (the layer becomes a filter of sorts), and using that layer as input to a second neural network that was optimizing a confusion loss.