These systems that leverage cycles also relates to newer research on ‘incremental learning’. One of the drawbacks of Deep Learning systems is that of the problem that ‘fine-tuning’ the network by training against new data can destroy previous remembered capabilities. This is the problem of the network ‘forgetting’ its past learning. In an architecture developed by Stanford called “Feedback Networks”, the researchers explored a different kind of network that feeds back into itself and develops the internal representation incrementally:
What about the other tech giants? There’s very little we know about their unannounced projects other than the fact that they’re all being worked on. With all cases, interest in Virtual Reality and Mixed Reality seem to go hand-in-hand, as both are the hottest new immersive experience mediums in the industry. Google has been spending considerable amounts of money in their Virtual Reality department and is also reportedly developing a Mixed Reality device of their own. Intel has announced Project Alloy, a VR/MR headset hybrid set to release next year. Apple has several patents for Mixed and Virtual Reality headsets, both of which should see the light of day sometime in 2017/2019. Additionally, Facebook and Samsung have created entirely new departments with hundreds of employees dedicated to VR and related technologies, spending billions of dollars in the process. Even SnapChat could be tiptoeing their way into MR as well, with a steady stream of cash going into computer vision (which currently powers their popular Snapchat filters) and a new pair of glasses with 115-degree cameras (The Snapchat’s Spectacles), which could very well be a way to test out the market’s reception to wearables of this kind.