Dynamic machine learning

for discovery services

How does it differ from conventional machine-learning?

  • The Xyggy Engine's automatically learns in realtime and so there is no accumulative re-training cost.

  • From Inside Facebook's Biggest AI Project Ever

  • “To do this Facebook run tens of trillions queries per day to make about six million predictions per second. Facebook trains the algorithms that power its News Feed within hours, using trillions of data points. The company updates its learning models every 15 minutes to two hours so that it can react quickly to current events.”

    The re-training of learning models is endemic to every machine learning system in the world and the cost in infrastructure, resources and time are enormous. It also tricks people into thinking that AI services are dynamic when in fact, they are static.

  • All data types are supported which includes text, images, audio and composite data.

  • A query is formed with one or more items and the Engine will find other relevant items in ranked order.

  • Deliver an interactive experience with flexible query options which is not possible with conventional machine learning (because it is not dynamic).

  • Automated end-to-end. Plug-in data at one end and query at the other, all with a simple API. Opens up powerful AI for developers with no previous machine learning experience.

  • For all types of discovery services including Personal Discovery, Autonomous Discovery, Intentional Discovery, RecSys, Feeds and so on.

Bayesian machine learning. Automatic learner. Deep learning. Distributed. Scalable. Interactive. API.

Email Hello Xyggy