Privacy and the Science of Data Analysis 2019
2019-04-08..2019-04-12
Berkeley, United States
| title | author | length |
|---|---|---|
| Locally Private Bayesian Inference for Count Models | Steven Wu | 48m |
Proposes locally private Bayesian inference for Poisson count models using the two-sided geometric mechanism for integer noise, rewrites noisy counts as… | ||
| Finite Sample Differentially Private Confidence Intervals | Vishesh Karwa | 47m |
Constructs finite-sample differentially private confidence intervals for a Gaussian mean (known and unknown variance) by first estimating a data range… | ||
| Differential Privacy with Elliptical Stochastic Processes | Matthew Reimherr | 40m |
Proposal: add noise drawn from an elliptical stochastic process to a function estimate, calibrate scale sigma by the RKHS norm of the change in the estimate… | ||
| Robust Statistics and Bias Correction: Ideas for Differential Privacy and Utility | Roberto Molinari | 48m |
Argues that using M-estimators with bounded score (Huber family, tuning K between 0.5–2) yields bounded sensitivity so less noise is needed under objective… | ||
| Differentially Private Inference for Binomial Data | Jordan Awan | 38m |
Derives the first differentially private uniformly most powerful tests for Bernoulli/binomial data under (ε,Δ)-DP and introduces the two-lap distribution (a… | ||
| The Structure of Optimal Private Tests for Simple Hypotheses | Audra McMillan | 48m |
Gives an instance-specific formula for the optimal sample size to privately distinguish two distributions: clamp each sample’s log‑likelihood ratio to ±epsilon… | ||
| Practical Difficulties in the Application of Differential Privacy for the Release of Large-Scale... | Philip LeClerc | 01h 00m |
Census builds a production disclosure avoidance system (DAS) that uses the high‑dimensional matrix mechanism and top‑down optimization to release microdata… | ||
| Shrinkwrap: Differentially-Private Query Processing in Private Data Federations | Johes Bater | 45m |
Shrinkwrap implements SMC-based query execution over horizontally partitioned hospital databases and adds differentially-private resizing to replace worst-case… | ||
| Analyzing Administrative Data with Privacy Protection in Place | Frauke Kreuter | 43m |
Administrative wage and employment records (nearly complete since 1972) require heavy episode-splitting and cross-source linkage (linkage handled by ~10 staff)… | ||
| Differentially Private Inference for Regression Coefficients | Andrés Felipe Barrientos | 44m |
Creates a synthetic version of the Office of Personnel Management federal employee dataset and builds a web verification server that answers regression queries… | ||
| Node Privacy, High-dimensional Graph Summaries, and Block Models | Adam M. Smith | 54m |
Algorithms for node-level differential privacy that estimate stochastic block model parameters and, in many regimes, match the optimal non-private error… | ||
| Optimality in Locally Private Estimation and Learning | John Duchi | 53m |
Proves lower bounds for locally private estimation by reducing estimation to testing via Assouad’s method and a communication-complexity / strong data… | ||
| PANEL: Statistical Theory, Privacy and Data Analysis | Aleksandra Slavkovic, Anand Sarwate, Thomas Steinke, Bin Yu, John Duchi, Yu-Xiang Wang | 01h 00m |
Panel argues that differential privacy acts as both a constraint and an enabler: it enforces data hygiene, gives formal generalization guarantees, and opens… | ||
| Differentially Private Change-Point Detection | Rachel Cummings | 47m |
Provides private algorithms for offline and online single and multiple change point detection that match nonprivate accuracy up to constants and add a privacy… | ||
| Bandit Learning under Differential Privacy | Or Sheffet | 44m |
Defines event-level pure differential privacy for bandit pulls, proves a lower bound that private algorithms must pull any suboptimal arm roughly Omega(log T… | ||
| Pufferfish Privacy Mechanisms for Correlated Data | Shuang Song | 39m |
Proposes Pufferfish privacy with two concrete mechanisms: a general Laplace-style method that sets noise by the infinity-Wasserstein distance between… | ||
| Black-box Random Differential Privacy with Sensitivity Sampling | Ben Rubinstein | 47m |
Paper proposes a sensitivity sampler that estimates the distribution of a function’s local sensitivities by drawing many i.i.d. samples, inverting the… | ||
| The Privacy Blanket of the Shuffle Model | Borja Balle | 42m |
Claims a single‑message shuffle protocol attains mean squared error Θ(n^{1/3}) (so estimation error Θ(n^{1/6})) with a matching lower bound, and achieves… | ||
| Open Problem Session | Shay Moran, Vitaly Feldman, Raef Bassily, Sofya Raskhodnikova, Thomas Steinke, Matthew Reimherr, Sashko Nikolov, Adam M. Smith, Adam Sealfon | 01h 19m |
Nine eight-minute open problems present concrete conjectures, partial results and explicit gaps: a private VC equivalence conjecture linking private uniform… | ||
| Locally Private Statistical Inference: Communication, Samples, and Shared Randomness | Jayadev Acharya | 44m |
Claims tight sample bounds for locally private inference: distribution learning needs about K^2/(ε^2 α^2) samples under LDP private‑coin (centralized is… | ||
| Privacy Amplification by Sampling and Renyi Differential Privacy | Yu-Xiang Wang | 50m |
Provides a usable amplification bound that maps a mechanism's Rényi DP curve to the subsampled mechanism, with near-tight constants and matching lower bounds… | ||
| Private Boosting and Noise-tolerant Learning | Mark Bun | 51m |
Constructs a private booster that enforces Sigma-smooth distributions by doing multiplicative-weights updates on an unnormalized measure and then… | ||
| Learning without Interaction Requires Separation | Vitaly Feldman | 36m |
Shows that non‑interactive local differential privacy (LDP) provably needs many more samples: any non‑interactive LDP PAC learner for a negation‑closed class… | ||
| Private Meta-Selection | Kunal Talwar | 44m |
Defines private meta‑selection: pick the best among K differentially private candidate mechanisms while avoiding an ε cost linear in K. Gives a randomized… | ||
| Private Center Points and Learning of Halfspaces | Amos Beimel | 46m |
Provides an (ε,δ)-private algorithm that finds an approximate center point (Tukey depth ≥ n/(2d+1)) of points on a finite grid and uses that to privately learn… | ||
| Federated Heavy Hitters Discovery with Differential Privacy | Peter Kairouz | 44m |
Federated algorithm builds a prefix tree by sampling M = M = γ√n devices each round, having each sampled device cast a single-character vote, and pruning… | ||
| Private Learning with Auxiliary Public Data | Raef Bassily | 53m |
A private predictor wraps any non‑private learner as a black-box transformation and uses subsample-and-aggregate / distance-to-instability plus the sparse… | ||
| Is Interaction Necessary for Distributed Private Learning? | Abhradeep Guha Thakurta | 43m |
Proves that local differential privacy forces a tradeoff between accuracy and rounds: one-dimensional convex ERM gets optimal O(1/√(n)ε) error with zero… | ||
| Groups and Individuals in Fairness and Probability | Cynthia Dwork | 53m |
Compares group fairness to individual fairness and argues individual fairness needs a task-specific metric; cites a metric-learning breakthrough that learns… | ||
| Towards a Characterization of Differentially Private Learning | Shay Moran | 39m |
Shows that private PAC learnability implies a finite Littlestone dimension and that private proper learning is equivalent to private uniform convergence… | ||
| Smooth Sensitivity, Revisited | Thomas Steinke | 47m |
A differentially private Gaussian mean estimator that matches the nonprivate variance up to a low‑order privacy cost, achieved by applying smooth sensitivity… | ||
| On the Power of Interactivity in Local Differential Privacy | Aaron Roth | 54m |
Defines a compositionality parameter K-compositional and proves any fully interactive ε‑DP protocol with that property compiles to a sequentially interactive… | ||
| The Limits of Post-Selection Generalization | Uri Stemmer | 36m |
Uri Stemmer (Ben Gurion University)
https://simons.berkeley.edu/talks/tba-66
Privacy and the Science of Data Analysis | ||
| A Necessary and Sufficient Stability Notion for Adaptive Generalization | Moshe Shenfeld | 46m |
Defines local statistical stability (LSS) as average total-variation between prior and posterior over a sampled point and proves LSS is both necessary and… | ||
| nothing matches | ||