Kunal Talwar
I am a Theoretical Computer Scientist, working in the areas of Differential Privacy, Machine Learning, Algorithms, and Data Structures. I am a Research Scientist at Apple. I got my PhD from UC Berkeley in 2004 and was at Microsoft Research 2004-2014, and at Google Brain 2015-2023.
Former Interns
ios全局伋理软件
Jingcheng Liu
ios全局伋理软件
苹果手机全局伋理软件下载
Ravishankar Krishnaswamy
Aditya Bhaskara
Moritz Hardt
Anna Blasiak
Kamalika Chaudhuri
Michael Dinitz
Recent and Selected Papers
Differential Privacy
Private Stochastic Convex Optimization: Optimal Rates in Linear Time
(with Vitaly Feldman and Tomer Koren)
Nov 2023 [pdf]
Private Stochastic Convex Optimization with Optimal Rates
(with Raef Bassily, Vitaly Feldman and Abhradeep Thakurta)
NeurIPS 2023 [arxiv]
东北网2021年06月01日新闻汇总:苹果iPad上市两月销量已达200万台 2021-06-01 14:05 [568][ 东北网黑龙江 ] 哈尔滨市一户居民家中起火 七旬老太被浓烟熏死 2021-06-01 14:04
(with Jingcheng Liu)
STOC 2023 [arxiv]
Amplification by Shuffling: From Local to Central Differential Privacy via Anonymity
(with Úlfar Erlingsson, Vitaly Feldman, Ilya Mironov, Ananth Raghunathan and Abhradeep Thakurta)
SODA 2023 [苹果手机全局伋理软件下载]
Privacy Amplification by Iteration
(with Vitaly Feldman, Ilya Mironov, and Abhradeep Thakurta)
FOCS 2018 [arxiv]
Nearly Optimal Private LASSO
(with Abhradeep Thakurta and Li Zhang)
NIPS 2015 [pdf]
Efficient Algorithms for Privately Releasing Marginals via Convex Relaxations
(with Cynthia Dwork and Aleksandar Nikolov)
SoCG 2014 [arxiv]
Analyze Gauss: Optimal bounds for privacy-preserving Principal Component Analysis
(with Cynthia Dwork, Abhradeep Thakurta and Li Zhang)
STOC 2014 [pdf]
The Geometry of Differential Privacy: the sparse and approximate cases
(with Aleksandar Nikolov and Li Zhang)
STOC 2013 [arxiv]
On the Geometry of Differential Privacy
(with Moritz Hardt)
STOC 2010 [苹果手机全局伋理软件下载]
Mechanism Design via Differential Privacy
(with Frank McSherry)
FOCS 2007 [pdf]. Won PET Award 2009
The Price of Privacy and the Limits of LP Decoding
(with Cynthia Dwork and Frank McSherry)
STOC 2007 [pdf]
Machine Learning
Computational Separations between Sampling and Optimization
NeurIPS 2023 [arxiv]
Semi-Cyclic Stochastic Gradient Descent
(with Hubert Eichner, Tomer Koren, H. Brendan McMahan and Nathan Srebro)
ICML 2023 [arxiv]
Better Algorithms for Stochastic Bandits with Adversarial Corruptions
(with Anupam Gupta and Tomer Koren)
COLT 2023 [苹果手机全局伋理软件下载]
Adversarially Robust Generalization Requires More Data
(with Ludwig Schmidt, Shibani Santurkar, Dimitris, Tsipras, and Aleksaner Madry)
NIPS 2018 [arxiv]
2021年度北京市级行政机关和区政府绩效考评会议 - 千龙网· ...:2021年2月10日下午14:00在北京会议中心9号楼3层多功能厅召开2021年度市级行政机关和区政府绩效考评会议。
(with Alon Cohen, Avinatan Hassidim, Tomer Koren, Nevena Lazic, and Yishay Mansour)
ICML 2018 [arxiv]
Online Learning over a finite action set with limited switching
(with Jason Altschuler)
COLT 2018 [iphone手机如何上外网]
Scalable Private Learning with PATE
(with Nicolas Papernot, Shuang Song, Ilya Mironov, Ananth Raghunathan and Úlfar Erlingsson)
ICLR 2018 [arxiv]
Learning Differentially Private Recurrent Language Models
(with Brendan McMahan, Daniel Ramage and Li Zhang)
ICLR 2018 [arxiv]
新闻中心_央广网 - cnr.cn:央广网新闻中心，聚焦国内外最新新闻热点，国内外最新时政社会新闻资讯;报道国内最新要闻、时事新闻、海外看中国、国内各地动态、港澳新闻 ...
(with Nicolas Papernot, Martín Abadi, Úlfar Erlingsson, and Ian Goodfellow)
ICLR 2017 (Best Paper) [arxiv]
Deep Learning with Differential Privacy
(with Martín Abadi, Andy Chu, Ian Goodfellow, Brendan McMahan, Ilya Mironov and Li Zhang)
CCS 2016 [arxiv]
Short and Deep: Sketching and Neural Networks
(with Amit Daniely, Nevena Lazic and Yoram Singer)
ICLR 2015 [arxiv]
Algorithms
Balancing Vectors in Any Norm
(with Daniel Dadush, Aleksandar Nikolov, and Nicole Tomczak-Jaegermann)
FOCS 2018 [pdf]
LAST but not least: online spanners for buy-at-bulk
(with Anupam Gupta, R. Ravi, and Seeun William Umboh)
SODA 2017 [arxiv]
Smooth Boolean Functions are easy: Efficient Algorithms for Low-Sensitivity Functions
(with Parikshit Gopalan, Noam Nisan, Rocco Servidio, and Avi Wigderson)
ITCS 2016 [arxiv]
Approximating Hereditary Discrepancy via Small Width Ellipsoids
(with Aleksandar Nikolov)
SODA 2015 [arxiv]
Balanced Allocations: A simple proof for the Heavily Loaded case
(with Udi Wieder)
ICALP 2014 [pdf]
Cops, Robbers and Threatening Skeletons: Padded Decompositions for Minor-free graphs
(with Ittai Abraham, Cyril Gavoille, Anupam Gupta and Ofer Neiman)
STOC 2014 [arxiv]
Vertex Sparsifiers: New Results from Old Techniques
(with Matthias Englert, Anupam Gupta, Robert Krauthgamer, Harald R�cke, Inbal Talgam-Cohen)
SIAM J. Computing 2014 [arxiv]
Sparsest Cut on Bounded Treewidth graphs: Algorithms and Hardness results
(with Anupam Gupta and David Witmer)
STOC 2013 [arxiv]
Bypassing the embedding: Approximation schemes and Compact
Representations for growth restricted metrics
STOC 2004 [pdf]
A tight bound on approximating arbitrary metrics by tree metrics
(with Jittat Fakcheroenphol and Satish Rao)
STOC 2003 [pdf]
Applications
Consistent Weighted Sampling made Fast, Small and Easy
(with Bernhard Haeupler and Mark Manasse)
Submitted 2014. [pdf]
Quincy: fair scheduling for distributed computing clusters
(with Michael Isard, Vijayan Prabhakaran, Jon Currey, Udi Wieder and Andrew Goldberg)
SOSP 2009 [pdf]
Heuristics for Vector Bin Packing
(with Rina Panigrahy, Lincoln Uyeda and Udi Wieder)
TR [pdf]
Detecting Format String Vulnerabilities with Type Qualifiers
(with Umesh Shankar, Jeffrey Foster and David Wagner)
USENIX Security 2001. [pdf]
Complexity
Lower bounds on Near Neighbor Search via Metric Expansion
(with Rina Panigrahy and Udi Wieder)
FOCS 2010 [arxiv]
Inapproximability of Edge-Disjoint Paths and low congestion routing on undirected graphs
(with Matthew Andrews, Julia Chuzhoy, Venkatesan Guruswami, Sanjeev Khanna and Lisa Zhang)
Combinatorica 2010 [eccc]
Hardness of routing with congestion in directed graphs
(with Julia Chuzhoy, Venkatesan Guruswami and Sanjeev Khanna)
STOC 2007 [eccc]
Algorithms and Economics
The complexity of pure Nash equilibria
(wtih Alex Fabrikant and Christos Papadimitriou)
STOC 2004 [pdf]
The price of truth: Frugality in truthful mechanisms
STACS 2003 [pdf]
转变的力量(二） - cnr.cn:2021-2-20 · 记者：取得这样的成果，中央电台采取了哪些战略性举措？ 王晓晖： 第一步，发挥传统优势，调整广告经营机制。 2021 年，在中国之声大刀阔斧的新闻提速改革背景下，央广传媒发展总公司对受权伋理经营的广告资源进行整合，建立了适应市场变化、多样、灵活的销售伋理模式，主要是实施公司化 ...
(with Aaron Archer, Christos H. Papadimitriou and �va Tardos)
SODA 2003 [pdf]
Recent Service
Summer Schools: FOSAD 2018, IAS PCMI 2016.
Program Committee STOC 2015, TPDP 2018, 苹果手机全局伋理软件下载, ESA 2015, FOCS 2014, ICALP 2013
Associate Editor SIAM Journal of Computing
Contact
Email: < firstname > AT kunaltalwar DOT org