Abstract
In this paper, we develop new test statistics for private hypothesis testing. These statistics are designed specifically so that their asymptotic distributions, after accounting for noise added for privacy concerns, match the asymptotics of the classical (nonprivate) chi-square tests for testing if the multinomial data parameters lie in lower dimensional manifolds (examples include goodness of fit and independence testing). Empirically, these new test statistics outperform prior work, which focused on noisy versions of existing statistics.
| Original language | English (US) |
|---|---|
| State | Published - Jan 1 2017 |
| Event | 20th International Conference on Artificial Intelligence and Statistics, AISTATS 2017 - Fort Lauderdale, United States Duration: Apr 20 2017 → Apr 22 2017 |
Conference
| Conference | 20th International Conference on Artificial Intelligence and Statistics, AISTATS 2017 |
|---|---|
| Country | United States |
| City | Fort Lauderdale |
| Period | 4/20/17 → 4/22/17 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Artificial Intelligence
- Statistics and Probability
Cite this
}
A new class of private chi-square tests. / Kifer, Daniel; Rogers, Ryan.
2017. Paper presented at 20th International Conference on Artificial Intelligence and Statistics, AISTATS 2017, Fort Lauderdale, United States.Research output: Contribution to conference › Paper
TY - CONF
T1 - A new class of private chi-square tests
AU - Kifer, Daniel
AU - Rogers, Ryan
PY - 2017/1/1
Y1 - 2017/1/1
N2 - In this paper, we develop new test statistics for private hypothesis testing. These statistics are designed specifically so that their asymptotic distributions, after accounting for noise added for privacy concerns, match the asymptotics of the classical (nonprivate) chi-square tests for testing if the multinomial data parameters lie in lower dimensional manifolds (examples include goodness of fit and independence testing). Empirically, these new test statistics outperform prior work, which focused on noisy versions of existing statistics.
AB - In this paper, we develop new test statistics for private hypothesis testing. These statistics are designed specifically so that their asymptotic distributions, after accounting for noise added for privacy concerns, match the asymptotics of the classical (nonprivate) chi-square tests for testing if the multinomial data parameters lie in lower dimensional manifolds (examples include goodness of fit and independence testing). Empirically, these new test statistics outperform prior work, which focused on noisy versions of existing statistics.
UR - http://www.scopus.com/inward/record.url?scp=85067546326&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85067546326&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:85067546326
ER -