Differential Privacy for Regularised Linear Regression
| dc.contributor.author | DANDEKAR, Ashish | |
| dc.contributor.author | BASU, Debabrota | |
| dc.contributor.author | BRESSAN, Stephane | |
| dc.date.accessioned | 2018-06-07T03:57:12Z | |
| dc.date.available | 2018-06-07T03:57:12Z | |
| dc.description.abstract | Recent attacks on machine learning models such as membership inference attacks increase the concern for privacy. Linear regression is such an essential statistical machine learning model at risk. For a given dataset, linear regression determines the parameters of the linear equation connecting the predictor variables to the response variable. As such linear regression yields a set of unstable and over tted parameters. Regularisation terms are added to the loss function of linear regression in order to avoid overftting. LASSO, ridge, and elastic net are three variants of regularised linear regression. We present an e-differentially private functional mechanism for the aforementioned variants of regularised linear regression. We empirically and comparatively analyze its effectiveness. A functional mechanism achieves differential privacy for linear regression by adding noise to the loss function. We empirically show that an e-differentially private functional mechanism causes more error than the non-private linear regression models whereas their performances are comparable. We also discuss caveats in the functional mechanism, such as non-convexity of the noisy loss function, which causes instability in the results of di erentially private linear regression models. This discussion puts forth the need of designing a differentially private mechanism that produces a noisy loss function that is convex. | en_US |
| dc.identifier.uri | https://dl.comp.nus.edu.sg/xmlui/handle/1900.100/7051 | |
| dc.relation.ispartofseries | TRB6/18 | en_US |
| dc.subject | linear regression, data privacy, differential privacy, elastic net | en_US |
| dc.title | Differential Privacy for Regularised Linear Regression | en_US |