If people could communicate with and interactively modify the behavior of AI systems, both people and machines could behave more intelligently. Unfortunately, most AI systems are black boxes designed to solve a single, narrowly defined problem, such as chess, face recognition, or click prediction, and adjusting their behavior requires deep technical expertise. Vikash Mansinghka and Richard Tibbetts describe the progress toward more transparent and flexible AI systems capable of augmenting rather than just replacing human intelligence, building on the emerging field of probabilistic programming, and explore how AI will be used on problems like malnutrition, public health, education, and governance—complex, ambiguous areas of human knowledge where data is sparse and there are no rules.
Probabilistic programming draws on probability theory, programming languages, and system software to provide concise, expressive languages for modeling and general-purpose inference engines that both humans and machines can use. Vikash and Richard focus on BayesDB and Picture, domain-specific probabilistic programming platforms aimed at augmenting intelligence in the fields of data science and computer vision, respectively. BayesDB, which is open source and in use by organizations like the Bill & Melinda Gates Foundation and JPMorgan, lets users who lack statistics training understand the probable implications of data by writing queries in a simple, SQL-like language. Picture, a probabilistic language being developed in collaboration with Microsoft, lets users solve hard computer vision problems such as inferring 3D models of faces, human bodies, and novel generic objects from single images by writing short (<50 line) computer graphics programs that generate and render random scenes. Unlike bottom-up vision algorithms, Picture programs build on prior knowledge about scene structure and produce complete 3D wireframes that people can manipulate using ordinary graphics software.
Vikash and Richard also briefly illustrate the fundamentals of probabilistic programming using Venture, an interactive platform suitable for teaching and for applications in fields ranging from statistics to robotics, and conclude with a summary of current and future research directions.
Vikash Mansinghka is a research scientist at MIT, where he leads the Probabilistic Computing Project, and a cofounder of Empirical Systems, a new venture-backed AI startup aimed at improving the credibility and transparency of statistical inference. Previously, Vikash cofounded a venture-backed startup based on his research that was acquired by Salesforce, was an advisor to Google DeepMind, and held graduate fellowships at the National Science Foundation and MIT’s Lincoln Laboratory. He served on DARPA’s Information Science and Technology advisory board from 2010 to 2012 and currently serves on the editorial boards for the Journal of Machine Learning Research and Statistics and Computation. Vikash holds a PhD in computation, an MEng in computer science, and BS degrees in mathematics and computer science, all from MIT. His PhD dissertation on natively probabilistic computation won the MIT George M. Sprowls dissertation award in computer science, and his research on the Picture probabilistic programming language won an award at CVPR.
Richard Tibbetts is currently a Principal Product Manager at Tableau. He was founder and CEO of Empirical Systems (acquired by Tableau 2018), a MIT spinout building an AI-based data platform that provided decision support to organizations that use structured data. Prior to Empirical, he was founder and CTO of StreamBase, a CEP company (acquired by TIBCO 2013), as well as a visiting scientist at the Probabilistic Computing Project at MIT.
©2016, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • email@example.com