21,461 Hits in 4.5 sec

Revisiting Differentially Private Hypothesis Tests for Categorical Data [article]

Yue Wang and Jaewoo Lee and Daniel Kifer
2017 arXiv   pre-print
Previous approaches to differentially private hypothesis testing either perturbed the test statistic with random noise having large variance (and resulted in a significant loss of power) or added smaller  ...  We then develop differentially private likelihood ratio and chi-squared tests for a variety of applications on tabular data (i.e., independence, sample proportions, and goodness-of-fit tests).  ...  Conclusions In this paper, we revisited the topic of -differentially private hypothesis testing. We A. Proof of Theorem 1 Theorem 1.  ... 
arXiv:1511.03376v4 fatcat:xcvqusq34bhaxh6zdc3ayng4de

Page 394 of Human Resources Abstracts Vol. 34, Issue 3 [page]

1999 Human Resources Abstracts  
This article tests the hypothesis that increased product market competition of deregulation decrease employer and union wage discrimination in the motor carrier industry.  ...  Wage compensation for dangerous work revisited. Industrial Labor Relations Review 52(1):116-135, Oct. 1998. DANGEROUS WORK COMPENSATION. PAY DIFFERENTIALS. WAGES.  ... 

Priv'IT: Private and Sample Efficient Identity Testing [article]

Bryan Cai, Constantinos Daskalakis, Gautam Kamath
2017 arXiv   pre-print
We develop differentially private hypothesis testing methods for the small sample regime.  ...  We experimentally compare the sample complexity of our method to that of recently proposed methods for private hypothesis testing.  ...  A different approach towards private hypothesis testing is to look deeper into the non-private tests and try to "privatize" them.  ... 
arXiv:1703.10127v3 fatcat:4dqnt3qlg5frpgoy3xvg337qye

The Influence of Destination Branding and Tourism Security on Revisit Intention During Covid-19 in the Uluwatu Temple, Bali, Indonesia

Jui-Lung Chen, Fellyana Yuni Indrianti
2022 International Journal of Business and Management  
The hypothesis tested with a purposive random sampling technique with the total number of 250 respondents.  ...  This study has six hypotheses and the results of all hypotheses support the intention to revisit and recommend.  ...  The steps in testing this hypothesis are started by establishing the null hypothesis (H 0 ) and the alternative hypothesis (H a ), selecting the statistical test and its calculation, determining the level  ... 
doi:10.5539/ijbm.v17n3p13 fatcat:zftihpga2jc2vl6jxfelxzqmya

Towards a Competitive Framework for Park Branding in Kenya: Case of Premium and Under-utilized Parks in Kenya

Shem Maingi Alice Ondigi
2014 Journal of Tourism Research & Hospitality  
Using the sample data, an analytical plan was instituted to test the null hypothesis.  ...  Hypothesis Two: Kenyan park brand perceptions do not significantly influence visitor revisit intentions In the second hypothesis, there was a specific need in assessing whether the park brand perception  ... 
doi:10.4172/2324-8807.1000141 fatcat:e7ctnvkgync4rdv26ygd6mge6e

The Role of Customer Experience, Food Healthiness, and Value for Revisit Intention in GROCERANT

Se Ran Yoo, Suk Won Lee, Hyeon Mo Jeon
2020 Sustainability  
that may enhance customer value and intention to revisit grocerants.  ...  This study differentiates itself from earlier studies of consumer experience in the foodservice industry that ignore the importance of food healthiness.  ...  Table 3 outlines the hypothesis test results that set the causal relationships between variables.  ... 
doi:10.3390/su12062359 fatcat:wn5js5e2abdvxnpnw7c4vijhum

Revisiting Membership Inference Under Realistic Assumptions [article]

Bargav Jayaraman, Lingxiao Wang, Katherine Knipmeyer, Quanquan Gu, David Evans
2021 arXiv   pre-print
Similarly to Liu et al. [2019] , we adopt a hypothesis testing perspective on differential privacy in which the adversary uses hypothesis testing on the differentially private mechanism's output to make  ...  Liu et al. [2019] theoretically evaluate differentially private mechanisms using a hypothesis testing framework using precision, recall and F-score metrics.  ...  Though the members and non-members are less differentiated, Morgan is still able to identify the most vulnerable members with 100% confidence (see Table 8 ). Results on CIFAR-100.  ... 
arXiv:2005.10881v5 fatcat:nlt2xp7bured3beiv3jfka22ve

Indonesian Favorite Destinations: Antecedent and Consequence Destination Experience

Nuryanti Taufik, Faizal Haris Eko Prabowo
2020 Jurnal Manajemen Teori dan Terapan | Journal of Theory and Applied Management  
The results found that revisit intention and e-WOM will be affected by destination experience.  ...  The attitude measurement scale is the differential magnetic scale, the primary data obtained from 502 respondents who had visited Bandung, or Yogyakarta, or Bali.  ...  Goodness of Fit Hypothesis Test Hypothesis testing is a step to test whether the research hypothesis can be accepted or rejected.  ... 
doi:10.20473/jmtt.v13i3.21239 fatcat:g7qa5yfhp5delnawbfi2k3t3vy

The Determinants of Visitor's Revisit Intention: A Lesson from Ijen Car Free Day

Cesya Rizkika Parahiyanti, Ananda Sabil Hussein
2014 APMBA (Asia Pacific Management and Business Application)  
It has a powerful function to make a differentiation between one destination and others.  ...  Keywords Event Brand Awareness (EBA), Event Brand Image (EBI), Event Brand Quality (EBQ), Event Revisit Intention (ERI)  ...  Therefore, Hypothesis 6 is not supported. The summary of Hypotheses testing is showed by table 3. Table 3.  ... 
doi:10.21776/ub.apmba.2014.003.02.1 fatcat:hc2xjrxz4vhm7kasteaaiqoz4y

A Statistical Threshold for Adversarial Classification in Laplace Mechanisms [article]

Ayşe Ünsal, Melek Önen
2021 arXiv   pre-print
Such trade-offs are provided for Laplace mechanisms using one-sided and two-sided hypothesis tests.  ...  An adversary who is able to modify the published information from a differentially private mechanism aims to maximize the possible damage to the system while remaining undetected.  ...  The authors wish to continue this line of work on different differentially private mechanisms (e.g.  ... 
arXiv:2105.05610v4 fatcat:2qxohjcst5ghvdvomwwesbc6vi

The Effect of Local Food on Tourism: Gaziantep Case

2019 Gaziantep University Journal of Social Sciences  
Local foods; which play an important role in destination choice, revisit intention and marketing are useful tools for destination development.  ...  It had been also found that the dimensions of cultural experience, variety and table manner, food taste have a positive effect on the revisit intention.  ...  test.  ... 
doi:10.21547/jss.560979 fatcat:zgc7bf7myvcz3lcblvhm4lmktm

Privately Answering Classification Queries in the Agnostic PAC Model [article]

Anupama Nandi, Raef Bassily
2019 arXiv   pre-print
We revisit the problem of differentially private release of classification queries.  ...  Given any hypothesis class with VC-dimension d, we show that our construction can privately answer up to m classification queries with average excess error α using a private sample of size ≈d/α^2 max(1  ...  Introduction In this paper, we revisit the problem of answering a sequence of classification queries in the agnostic PAC model under the constraint of (ǫ, δ)-differential privacy.  ... 
arXiv:1907.13553v3 fatcat:vtzan4kxwngbnmzfdxhyqdiloe

Revisiting Model-Agnostic Private Learning: Faster Rates and Active Learning [article]

Chong Liu, Yuqing Zhu, Kamalika Chaudhuri, Yu-Xiang Wang
2022 arXiv   pre-print
The Private Aggregation of Teacher Ensembles (PATE) framework is one of the most promising recent approaches in differentially private learning.  ...  Revisiting Model-Agnostic Private Learning: Faster Rates and Active Learning  ...  Differential Privacy and Private Learning Now we formally introduce differential privacy.  ... 
arXiv:2011.03186v4 fatcat:gdybamtybzfqhdw6szh7ebxaci

Bolt-on Differential Privacy for Scalable Stochastic Gradient Descent-based Analytics [article]

Xi Wu, Fengan Li, Arun Kumar, Kamalika Chaudhuri, Somesh Jha, Jeffrey F. Naughton
2017 arXiv   pre-print
differentially private SGD.  ...  In contrast to the white-box approach adopted by previous work, we revisit and use the classical technique of output perturbation to devise a novel "bolt-on" approach to private SGD.  ...  Algorithm 2 is ε-differentially private.  ... 
arXiv:1606.04722v3 fatcat:5pz2wqw5tbdzfgwlm7kek74zdi

Bolt-on Differential Privacy for Scalable Stochastic Gradient Descent-based Analytics

Xi Wu, Fengan Li, Arun Kumar, Kamalika Chaudhuri, Somesh Jha, Jeffrey Naughton
2017 Proceedings of the 2017 ACM International Conference on Management of Data - SIGMOD '17  
differentially private SGD.  ...  In contrast to the white-box approach adopted by previous work, we revisit and use the classical technique of output perturbation to devise a novel "bolt-on" approach to private SGD.  ...  have incorporated differentially private SGD.  ... 
doi:10.1145/3035918.3064047 dblp:conf/sigmod/0001LKCJN17 fatcat:7nqou7wh4rfn5jrylmgc6olqu4
« Previous Showing results 1 — 15 out of 21,461 results