<ul><li>This research paper introduces a method called DP-IHT-H for differentially private sparse linear regression with heavy-tailed responses in high-dimensional settings.</li><li>DP-IHT-H leverages the Huber loss and private iterative hard thresholding to achieve an estimation error bound under the differential privacy model.</li><li>Another method proposed in the paper, DP-IHT-L, further improves the error bound under additional assumptions on the response and achieves better results.</li><li>Experiments conducted on synthetic and real-world datasets show that these methods outperform standard differentially private algorithms designed for regular data.</li></ul>

Differentially Private Sparse Linear Regression with Heavy-tailed Responses

Discover more