Share This Author
Matrix Completion From a Few Entries
- Raghunandan H. Keshavan, A. Montanari, Sewoong Oh
- Mathematics, Computer ScienceIEEE Transactions on Information Theory
- 20 January 2009
An efficient algorithm is described, which is called OptSpace, that reconstructs M from |E| = O(rn) observed entries with relative root mean square error 1/2 RMSE ¿ C(¿) (nr/|E|)1/2 with probability larger than 1 - 1/n3.
Matrix Completion from Noisy Entries
This work studies a low complexity algorithm, introduced in , based on a combination of spectral techniques and manifold optimization, that is called here OPTSPACE, and proves performance guarantees that are order-optimal in a number of circumstances.
Extremal Mechanisms for Local Differential Privacy
It is shown that for all information theoretic utility functions studied in this paper, maximizing utility is equivalent to solving a linear program, the outcome of which is the optimal staircase mechanism, which is universally optimal in the high and low privacy regimes.
PacGAN: The Power of Two Samples in Generative Adversarial Networks
- Zinan Lin, Ashish Khetan, G. Fanti, Sewoong Oh
- Computer ScienceIEEE Journal on Selected Areas in Information…
- 12 December 2017
It is shown that packing naturally penalizes generators with mode collapse, thereby favoring generator distributions with less mode collapse during the training process, and numerical experiments suggests that packing provides significant improvements in practice as well.
The Composition Theorem for Differential Privacy
- P. Kairouz, Sewoong Oh, P. Viswanath
- Computer Science, MathematicsIEEE Transactions on Information Theory
- 4 November 2013
This paper proves an upper bound on the overall privacy level and construct a sequence of privatization mechanisms that achieves this bound by introducing an operational interpretation of differential privacy and the use of a data processing inequality.
Rank Centrality: Ranking from Pairwise Comparisons
Experimental evaluations on synthetic data sets generated according to the popular Bradley-Terry-Luce (BTL) model show that the proposed Rank Centrality algorithm performs as well as the maximum likelihood estimator for that model and outperforms other popular ranking algorithms.
Attention-based Graph Neural Network for Semi-supervised Learning
A novel graph neural network is proposed that removes all the intermediate fully-connected layers, and replaces the propagation layers with attention mechanisms that respect the structure of the graph, and demonstrates that this approach outperforms competing methods on benchmark citation networks datasets.
Iterative ranking from pair-wise comparisons
This paper proposes a novel iterative rank aggregation algorithm for discovering scores for objects from pairwise comparisons which performs as well as the Maximum Likelihood Estimator of the BTL model and outperforms a recently proposed algorithm by Ammar and Shah.
Iterative Learning for Reliable Crowdsourcing Systems
This paper gives a new algorithm for deciding which tasks to assign to which workers and for inferring correct answers from the workers' answers, and shows that the algorithm significantly outperforms majority voting and is asymptotically optimal through comparison to an oracle that knows the reliability of every worker.
Budget-Optimal Task Allocation for Reliable Crowdsourcing Systems
A new algorithm is given for deciding which tasks to assign to which workers and for inferring correct answers from the workers' answers, and it is shown that the minimum price necessary to achieve a target reliability scales in the same manner under both adaptive and nonadaptive scenarios.