Zobrazeno 1 - 10
of 48
pro vyhledávání: '"Durfee, David"'
Autor:
Durfee, David
In this work we consider the problem of differentially private computation of quantiles for the data, especially the highest quantiles such as maximum, but with an unbounded range for the dataset. We show that this can be done efficiently through a s
Externí odkaz:
http://arxiv.org/abs/2305.01177
Autor:
Behdin, Kayhan, Song, Qingquan, Gupta, Aman, Keerthi, Sathiya, Acharya, Ayan, Ocejo, Borja, Dexter, Gregory, Khanna, Rajiv, Durfee, David, Mazumder, Rahul
Modern deep learning models are over-parameterized, where different optima can result in widely varying generalization performance. The Sharpness-Aware Minimization (SAM) technique modifies the fundamental loss function that steers gradient descent m
Externí odkaz:
http://arxiv.org/abs/2302.09693
Autor:
Behdin, Kayhan, Song, Qingquan, Gupta, Aman, Durfee, David, Acharya, Ayan, Keerthi, Sathiya, Mazumder, Rahul
Modern deep learning models are over-parameterized, where the optimization setup strongly affects the generalization performance. A key element of reliable optimization for these systems is the modification of the loss function. Sharpness-Aware Minim
Externí odkaz:
http://arxiv.org/abs/2212.04343
We introduce the notion of heterogeneous calibration that applies a post-hoc model-agnostic transformation to model outputs for improving AUC performance on binary classification tasks. We consider overconfident models, whose performance is significa
Externí odkaz:
http://arxiv.org/abs/2202.04837
Autor:
Rogers, Ryan, Subramaniam, Subbu, Peng, Sean, Durfee, David, Lee, Seunghyun, Kancha, Santosh Kumar, Sahay, Shraddha, Ahammad, Parvez
We present a privacy system that leverages differential privacy to protect LinkedIn members' data while also providing audience engagement insights to enable marketing analytics related applications. We detail the differentially private algorithms an
Externí odkaz:
http://arxiv.org/abs/2002.05839
Composition is one of the most important properties of differential privacy (DP), as it allows algorithm designers to build complex private algorithms from DP primitives. We consider precise composition bounds of the overall privacy loss for exponent
Externí odkaz:
http://arxiv.org/abs/1909.13830
Autor:
Durfee, David, Dhulipala, Laxman, Kulkarni, Janardhan, Peng, Richard, Sawlani, Saurabh, Sun, Xiaorui
In this paper we study the problem of dynamically maintaining graph properties under batches of edge insertions and deletions in the massively parallel model of computation. In this setting, the graph is stored on a number of machines, each having sp
Externí odkaz:
http://arxiv.org/abs/1908.01956
We study \emph{dynamic} algorithms for maintaining spectral vertex sparsifiers of graphs with respect to a set of terminals $T$ of our choice. Such objects preserve pairwise resistances, solutions to systems of linear equations, and energy of electri
Externí odkaz:
http://arxiv.org/abs/1906.10530
Autor:
Durfee, David, Rogers, Ryan
We study the problem of top-$k$ selection over a large domain universe subject to user-level differential privacy. Typically, the exponential mechanism or report noisy max are the algorithms used to solve this problem. However, these algorithms requi
Externí odkaz:
http://arxiv.org/abs/1905.04273
We give an algorithm to compute a one-dimensional shape-constrained function that best fits given data in weighted-$L_{\infty}$ norm. We give a single algorithm that works for a variety of commonly studied shape constraints including monotonicity, Li
Externí odkaz:
http://arxiv.org/abs/1905.02149