Matches in SemOpenAlex for { <https://semopenalex.org/work/W3136559452> ?p ?o ?g. }
- W3136559452 abstract "We propose a general optimization-based framework for computing differentially private M-estimators and a new method for constructing differentially private confidence regions. Firstly, we show that robust statistics can be used in conjunction with noisy gradient descent or noisy Newton methods in order to obtain optimal private estimators with global linear or quadratic convergence, respectively. We establish local and global convergence guarantees, under both local strong convexity and self-concordance, showing that our private estimators converge with high probability to a nearly optimal neighborhood of the non-private M-estimators. Secondly, we tackle the problem of parametric inference by constructing differentially private estimators of the asymptotic variance of our private M-estimators. This naturally leads to approximate pivotal statistics for constructing confidence regions and conducting hypothesis testing. We demonstrate the effectiveness of a bias correction that leads to enhanced small-sample empirical performance in simulations. We illustrate the benefits of our methods in several numerical examples." @default.
- W3136559452 created "2021-03-29" @default.
- W3136559452 creator A5008148805 @default.
- W3136559452 creator A5017137031 @default.
- W3136559452 creator A5023368186 @default.
- W3136559452 date "2021-03-19" @default.
- W3136559452 modified "2023-09-26" @default.
- W3136559452 title "Differentially private inference via noisy optimization." @default.
- W3136559452 cites W112690700 @default.
- W3136559452 cites W1886087434 @default.
- W3136559452 cites W1985511977 @default.
- W3136559452 cites W1992926795 @default.
- W3136559452 cites W1997859853 @default.
- W3136559452 cites W2008789956 @default.
- W3136559452 cites W2012712694 @default.
- W3136559452 cites W2018038591 @default.
- W3136559452 cites W2027595342 @default.
- W3136559452 cites W2045744861 @default.
- W3136559452 cites W2046033161 @default.
- W3136559452 cites W2064301386 @default.
- W3136559452 cites W2074006684 @default.
- W3136559452 cites W2074320615 @default.
- W3136559452 cites W2099756352 @default.
- W3136559452 cites W2112380340 @default.
- W3136559452 cites W2119874464 @default.
- W3136559452 cites W2121210949 @default.
- W3136559452 cites W2124768887 @default.
- W3136559452 cites W2127555295 @default.
- W3136559452 cites W2130486630 @default.
- W3136559452 cites W2184139426 @default.
- W3136559452 cites W2186926886 @default.
- W3136559452 cites W2296319761 @default.
- W3136559452 cites W2473418344 @default.
- W3136559452 cites W2603538444 @default.
- W3136559452 cites W2616050959 @default.
- W3136559452 cites W2680102885 @default.
- W3136559452 cites W2751484150 @default.
- W3136559452 cites W2807215287 @default.
- W3136559452 cites W2809607392 @default.
- W3136559452 cites W2916987143 @default.
- W3136559452 cites W2932329902 @default.
- W3136559452 cites W2962684583 @default.
- W3136559452 cites W2962727083 @default.
- W3136559452 cites W2962961705 @default.
- W3136559452 cites W2963038732 @default.
- W3136559452 cites W2963326510 @default.
- W3136559452 cites W2963559079 @default.
- W3136559452 cites W2963695762 @default.
- W3136559452 cites W2963776003 @default.
- W3136559452 cites W2963881987 @default.
- W3136559452 cites W2963927498 @default.
- W3136559452 cites W2963981973 @default.
- W3136559452 cites W2964064173 @default.
- W3136559452 cites W2964089577 @default.
- W3136559452 cites W2964161869 @default.
- W3136559452 cites W2964244224 @default.
- W3136559452 cites W2964303576 @default.
- W3136559452 cites W2971307541 @default.
- W3136559452 cites W2975597024 @default.
- W3136559452 cites W2993782690 @default.
- W3136559452 cites W3034801477 @default.
- W3136559452 cites W3046518446 @default.
- W3136559452 cites W3101914454 @default.
- W3136559452 cites W3102303514 @default.
- W3136559452 cites W3120817348 @default.
- W3136559452 cites W3143451908 @default.
- W3136559452 cites W3169993745 @default.
- W3136559452 cites W3189333525 @default.
- W3136559452 cites W3195182508 @default.
- W3136559452 cites W3198348214 @default.
- W3136559452 cites W3208102029 @default.
- W3136559452 cites W3211357969 @default.
- W3136559452 cites W44394327 @default.
- W3136559452 hasPublicationYear "2021" @default.
- W3136559452 type Work @default.
- W3136559452 sameAs 3136559452 @default.
- W3136559452 citedByCount "2" @default.
- W3136559452 countsByYear W31365594522021 @default.
- W3136559452 crossrefType "posted-content" @default.
- W3136559452 hasAuthorship W3136559452A5008148805 @default.
- W3136559452 hasAuthorship W3136559452A5017137031 @default.
- W3136559452 hasAuthorship W3136559452A5023368186 @default.
- W3136559452 hasConcept C105795698 @default.
- W3136559452 hasConcept C106159729 @default.
- W3136559452 hasConcept C117251300 @default.
- W3136559452 hasConcept C121955636 @default.
- W3136559452 hasConcept C126255220 @default.
- W3136559452 hasConcept C129844170 @default.
- W3136559452 hasConcept C144133560 @default.
- W3136559452 hasConcept C154945302 @default.
- W3136559452 hasConcept C162324750 @default.
- W3136559452 hasConcept C185429906 @default.
- W3136559452 hasConcept C196083921 @default.
- W3136559452 hasConcept C2524010 @default.
- W3136559452 hasConcept C26517878 @default.
- W3136559452 hasConcept C2776214188 @default.
- W3136559452 hasConcept C2777303404 @default.
- W3136559452 hasConcept C28826006 @default.
- W3136559452 hasConcept C33923547 @default.
- W3136559452 hasConcept C38652104 @default.