Timezone: »

Privacy-preserving logistic regression
Kamalika Chaudhuri · Claire Monteleoni

Mon Dec 08 08:45 PM -- 12:00 AM (PST) @

This paper addresses the important tradeoff between privacy and learnability, when designing algorithms for learning from private databases. First we apply an idea of Dwork et al. to design a specific privacy-preserving machine learning algorithm, logistic regression. This involves bounding the sensitivity of logistic regression, and perturbing the learned classifier with noise proportional to the sensitivity. Noting that the approach of Dwork et al. has limitations when applied to other machine learning algorithms, we then present another privacy-preserving logistic regression algorithm. The algorithm is based on solving a perturbed objective, and does not depend on the sensitivity. We prove that our algorithm preserves privacy in the model due to Dwork et al., and we provide a learning performance guarantee. Our work also reveals an interesting connection between regularization and privacy.

Author Information

Kamalika Chaudhuri (UCSD)
Claire Monteleoni (University of Colorado Boulder)

Claire Monteleoni is an associate professor of Computer Science at University of Colorado Boulder. Previously, she was an associate professor at George Washington University, and research faculty at the Center for Computational Learning Systems, at Columbia University. She did a postdoc in Computer Science and Engineering at the University of California, San Diego, and completed her PhD and Masters in Computer Science, at MIT. She holds a Bachelors in Earth and Planetary Sciences from Harvard. Her research focuses on machine learning algorithms and theory for problems including learning from data streams, learning from raw (unlabeled) data, learning from private data, and climate informatics: accelerating discovery in climate science with machine learning. Her work on climate informatics received the Best Application Paper Award at NASA CIDU 2010. In 2011, she co-founded the International Workshop on Climate Informatics, which is now in its fourth year, attracting climate scientists and data scientists from over 14 countries and 26 states.

More from the Same Authors