We’ve been in stealth mode for the past year creating, tinkering away at a new project. Throughout 2016 we’ve been building out our AI Labs In Amherst, Massachusetts—on the East bank of the Connecticut River. We’re calling the project Magic Machines. The goal is force-multiplying AI technologies. What does this mean? Well, the breakthrough technologies that comprise Magic Machines AI Labs address the severe limitations imposed by the complexity of machine learning.
The product is a highly-flexible, self-organizing, self-learning system that will work with humans to produce specific AI solutions to business problems. No matter your experience, no matter your business—Magic Machines will put the full power of the AI revolution in your hands. Over the coming months, Lexalytics, an InMoment company, will announce products that enable the average business person to handle more projects like never before. No longer are you limited to the functionality and industry focus out of the box. Magic Machines empower business intelligence users to shape the AI with simple, natural language commands.
We’ve partnered with the University of Massachusetts Amherst’s Center for Data Science and Northwestern University’s Medill School of Journalism, Media and Integrated Marketing Communications to create a powerful, beautiful, and effective solution. Through our affiliation with UMass, we’re working with faculty and students on the underlying challenges of analyzing, visualizing and drawing insights from massive volumes and varieties of data. Our partners at Northwestern are working with us to ensure the usability and applicability of Magic Machines AI technologies to a broad set of business users.
We’ve decided to force multiply five technologies, each selected to optimize the automation of building an application-specific AI: Swarm Intelligence/Emergent Behavior, Adaptive AI, Algorithmic Coordination, Transfer Learning and Meta-Learning. This foundational tech will allow data scientists to apply their time more efficiently while making sure business analysts can shape the resultant AI systems. Suffice it to say, this was not easy! In a recent Datanami article, our Chief Scientist, Paul Barba, put it this way: “Text analytics is an AI-hard problem, meaning that we have solved some of the most difficult AI-related challenges around. After over a year of working quietly on Magic Machines technologies, we’re happy with our results, and ready to catapult innovation in AI even further.”
So, the road to the marketplace was far from smooth. But through that we’ve overcome some of the most substantial and sophisticated AI problems. And now we’re ready to share our findings with the world.