Felix Zhou
LightNews LightNews
felix-zhou-cfz.bsky.social
Felix Zhou
@felix-zhou-cfz.bsky.social
24 followers 100 following 9 posts
CS PhD student @Yale

Algorithms, Differential Privacy, Machine Learning

http://felix-zhou.com
Posts Replies Media Videos
Pinned
felix-zhou-cfz.bsky.social
Felix Zhou @felix-zhou-cfz.bsky.social · Oct 18
Previous densest subgraph algorithms in the continual release model incur O(log n) error and space overhead compared to static counterparts.
By densifying the input graph, this overhead can be removed!

To appear in SOSA'26.
dppapers.bsky.social Differential Privacy Papers @dppapers.bsky.social · Oct 14
Continual Release of Densest Subgraphs: Privacy Amplification & Sublinear Space via Subsampling

Felix Zhou

http://arxiv.org/abs/2510.11640
Continual Release of Densest Subgraphs: Privacy Amplification & Sublinear Space via Subsampling Felix Zhou http://arxiv.org/abs/2510.11640 We study the sublinear space continual release model for edge-differentially private (DP) graph algorithms, with a focus on the densest subgraph problem (DSG) in the insertion-only setting. Our main result is the first continual release DSG algorithm that matches the additive error of the best static DP algorithms and the space complexity of the best non-private streaming algorithms, up to constants. The key idea is a refined use of subsampling that simultaneously achieves privacy amplification and sparsification, a connection not previously formalized in graph DP. Via a simple black-box reduction to the static setting, we obtain both pure and approximate-DP algorithms with $O(\log n)$ additive error and $O(n\log n)$ space, improving both accuracy and space complexity over the previous state of the art. Along the way, we introduce graph densification in the graph DP setting, adding edges to trigger earlier subsampling, which removes the extra logarithmic factors in error and space incurred by prior work [ELMZ25]. We believe this simple idea may be of independent interest.
felix-zhou-cfz.bsky.social
Previous densest subgraph algorithms in the continual release model incur O(log n) error and space overhead compared to static counterparts.
By densifying the input graph, this overhead can be removed!

To appear in SOSA'26.
dppapers.bsky.social Differential Privacy Papers @dppapers.bsky.social · Oct 14
Continual Release of Densest Subgraphs: Privacy Amplification & Sublinear Space via Subsampling

Felix Zhou

http://arxiv.org/abs/2510.11640
Continual Release of Densest Subgraphs: Privacy Amplification & Sublinear Space via Subsampling Felix Zhou http://arxiv.org/abs/2510.11640 We study the sublinear space continual release model for edge-differentially private (DP) graph algorithms, with a focus on the densest subgraph problem (DSG) in the insertion-only setting. Our main result is the first continual release DSG algorithm that matches the additive error of the best static DP algorithms and the space complexity of the best non-private streaming algorithms, up to constants. The key idea is a refined use of subsampling that simultaneously achieves privacy amplification and sparsification, a connection not previously formalized in graph DP. Via a simple black-box reduction to the static setting, we obtain both pure and approximate-DP algorithms with $O(\log n)$ additive error and $O(n\log n)$ space, improving both accuracy and space complexity over the previous state of the art. Along the way, we introduce graph densification in the graph DP setting, adding edges to trigger earlier subsampling, which removes the extra logarithmic factors in error and space incurred by prior work [ELMZ25]. We believe this simple idea may be of independent interest.
October 18, 2025 at 5:21 PM
felix-zhou-cfz.bsky.social
"What makes a good fisherman as opposed to other professions?"
This question can be formulated as a k-linear regression problem with self-selection bias.

Alkis, @anaymehrotra.bsky.social, and I design faster local convergence algorithms for this problem:
arxiv.org/abs/2504.07133

(1/7)
Can SGD Select Good Fishermen? Local Convergence under Self-Selection Biases and Beyond
We revisit the problem of estimating $k$ linear regressors with self-selection bias in $d$ dimensions with the maximum selection criterion, as introduced by Cherapanamjeri, Daskalakis, Ilyas, and Zamp...
arxiv.org
April 19, 2025 at 5:39 PM