Scientists from Charles River Analytics, a developer of intelligent systems solutions, presented the Figaro probabilistic programming language at the 18th International Conference on Information Fusion. The conference took place from July 6-9, 2015, at the Grand Hyatt Hotel in Washington DC.
Charles River presented Reasoning on Resident Space Object Hierarchies Using Probabilistic Programming, a joint effort with Applied Defense Solutions. Dr. Brian Ruttenberg, a senior scientist at Charles River Analytics, explained, “The impact of a resident space object, or RSO, with a satellite can be catastrophic, yet preventing this occurrence is difficult due the number and size of RSOs in space. Recent advances have proposed using a hierarchy or taxonomy to classify RSOs, to improve accuracy and triage detection methods. Representing and reasoning on these hierarchies is difficult, however, as they can be large, deep, uncertain, and constantly evolving.”
“We presented our solution to reasoning on probabilistic hierarchies for RSO classification using Figaro, a probabilistic programming language,” continued Dr. Ruttenberg. “Probabilistic programming provides the programming language and artificial intelligence tools needed to build and reason on large and evolving hierarchies of RSOs.”