Those focal areas include deep-learning algorithms that can help neural networks move from single-use applications to more generalized performance. Not only will this make AI systems more versatile, it will improve their transparency as well, enabling them to explain how they reached the answer they did.
The IBM-MIT partnership will also study the intersection between machine learning and quantum computing. Interestingly, this focal area will aid both fields, with AI helping to identify and characterize quantum devices and with quantum computers helping to optimize machine learning methodologies.
The MIT lab will also collaborate extensively with the IBM Watson Health and Security office in nearby Cambridge, Massachusetts, to further develop applications in the existing AI healthcare and cybersecurity fields. But those aren't the only commercial fields being investigated, researchers will also look into the "economic implications of AI and investigate how AI can improve prosperity," according to an IBM press release.
This isn't the first time that these organizations have worked together. Just last year, IBM and the MIT Department of Brain and Cognitive Sciences began orchestrating a machine vision study. What's more, IBM has also teamed with MIT's Broad Institute and Harvard for a multi-year study of AI's effects on Genomics.