Our research focuses on theory and algorithms for extracting actionable information from complex data streams. We are particularly interested in decision-making under uncertainty for autonomy in a variety of application areas. SLI is led by John Fisher and is part of the Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology.
Methods: We develop scalable and robust methods in Bayesian inference, information theory, optimization and machine learning. Of particular interest are Bayesian nonparametrics, information planning, scene understanding, and Bayesian causal inference.
Applications: Applications expose our ideas to the complexity of real world. We collaborate with researchers across the institute and industry partners on problems including materials discovery, animal behavior analysis, autonomous systems, and nuclear materials detection.
Sensors: Physics-based sensor models provide robustness and accurate uncertainty quantification in high-stakes sensing applications. We utilize a variety (and growing list) of sensor modalities – cameras, lidars, RGBd, particle detectors, application-specific sensors – within our methods.