Value of Information Lab

This lab is devising methods that reason about the acquisition cost and value of specific pieces of information when learning and making inferences with predictive models. These methods are being applied to determine how experimental, clinical, human and other resources should be optimally allocated when gathering data for training predictive models, and also to determine what information should be collected at prediction time in order to optimally make decisions using a given model.

Related CPCP Publications

Hawkes process modeling of adverse drug reactions with longitudinal observational data. Bao Y, Kuang Z, Peissig P, Page D, Willett R. Proceedings of Machine Learning for Healthcare, 2017

Publication details

Machine learning consensus scoring improves performance across targets in structure-based virtual screening. Ericksen S, Wu H, Zhang H, Michael L, Newton M, Hoffmann FM, Wildman S. Journal of Chemical Information and Modeling 57(7):1579–1590, 2017

Publication details

A review of active learning approaches to experimental design for uncovering biological networks. Sverchkov Y, Craven M. PLoS Computational Biology, 2017

Publication details

Adaptive signal recovery on graphs via harmonic analysis for experimental design in neuroimaging. Kim WH, Hwang SJ, Adluru N, Johnson SC, Singh V. Proceedings of the 14th European Conference on Computer Vision (ECCV), Volume 9910 Lecture Notes in Computer Science, 2016

Publication details

Experimental design on a budget for sparse linear models and applications. Ravi SN, Ithapu VK, Johnson SC, Singh V. Proceedings of the 33rd International Conference on Machine Learning (ICML), 2016

Publication details


Rob Nowak


Rebecca Willett

Aritra Biswas

Ravi Sastry Ganti Mahapatruni

Kwang-Sung Jun


CPCP Retreat 2016: Multi-Armed Bandit Algorithms and Applications to Experiment Selection Symposium Video

CPCP Retreat 2016: Using Active Learning to Phenotype Electronic Medical Records Symposium Video