Large Margin Multiclass Gaussian Classification with Differential Privacy

Pathak, Manas A.; Raj, Bhiksha

doi:10.1007/978-3-642-19896-0_9

Cited by 6 publications

(4 citation statements)

References 14 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance (i.e., privacy and utility) of a differentially private method is highly dependent on the nature of the application and the capability of the protection mechanism. To meet the need of different applications, many customized differentially private methods, including decision trees [ 10 ], logistic regression [ 11 ], principal components analysis [ 12 ], multi-class Gaussian classifiers [ 13 ], have been developed. There are several recent efforts in integrating the differential privacy framework into the system design and case studies for statistical health information release [ 14 ], [ 15 ].…”

Section: Introductionmentioning

confidence: 99%

Differentially private distributed logistic regression using private and public data

Jiang

Wang

et al. 2014

BMC Med Genomics

View full text Add to dashboard Cite

BackgroundPrivacy protecting is an important issue in medical informatics and differential privacy is a state-of-the-art framework for data privacy research. Differential privacy offers provable privacy against attackers who have auxiliary information, and can be applied to data mining models (for example, logistic regression). However, differentially private methods sometimes introduce too much noise and make outputs less useful. Given available public data in medical research (e.g. from patients who sign open-consent agreements), we can design algorithms that use both public and private data sets to decrease the amount of noise that is introduced.MethodologyIn this paper, we modify the update step in Newton-Raphson method to propose a differentially private distributed logistic regression model based on both public and private data.Experiments and resultsWe try our algorithm on three different data sets, and show its advantage over: (1) a logistic regression model based solely on public data, and (2) a differentially private distributed logistic regression model based on private data under various scenarios.ConclusionLogistic regression models built with our new algorithm based on both private and public datasets demonstrate better utility than models that trained on private or public datasets alone without sacrificing the rigorous privacy guarantee.

show abstract

Section: Introductionmentioning

confidence: 99%

Differentially private distributed logistic regression using private and public data

Jiang

Wang

et al. 2014

BMC Med Genomics

View full text Add to dashboard Cite

show abstract

“…Among words belonging to the DP community, 'multiclass' is interesting. After a small number of papers in the early 2010's [20,24], multiclass problems in DP have started to receive more attention recently [25]. Words in the non-DP cluster, on the other hand, refer to methods or algorithms-'mdps' is Markov Decision Processes, 'vb' is variational bayes, and 'triplet' is triplet loss.…”

Section: Discussionmentioning

confidence: 99%

An Interpretable Graph-based Mapping of Trustworthy Machine Learning Research

Derzsy,

Majumdar,

Malik

2021

Preprint

View full text Add to dashboard Cite

There is an increasing interest in ensuring machine learning (ML) frameworks behave in a socially responsible manner and are deemed trustworthy. Although considerable progress has been made in the field of Trustworthy ML (TwML) in the recent past, much of the current characterization of this progress is qualitative. Consequently, decisions about how to address issues of trustworthiness and future research goals are often left to the interested researcher. In this paper, we present the first quantitative approach to characterize the comprehension of TwML research. We build a co-occurrence network of words using a web-scraped corpus of more than 7,000 peer-reviewed recent ML papers-consisting of papers both related and unrelated to TwML. We use community detection to obtain semantic clusters of words in this network that can infer relative positions of TwML topics. We propose an innovative fingerprinting algorithm to obtain probabilistic similarity scores for individual words, then combine them to give a paper-level relevance score. The outcomes of our analysis inform a number of interesting insights on advancing the field of TwML research.

show abstract

“…At the same time, in order to solve the problem that the DiffP-ID3 algorithm can only deal with the limitation of discrete attributes, Friedman and Schuster extended the algorithm to continuous attributes and proposed the DiffP-C4.5 algorithm [8]. In 2010, Pathak and Raj proposed a large multi-category Gaussian classification based on differential privacy protection [9]. The classifier uses a large loss function with a perturbation regular term to satisfy the differential privacy, and gives the theoretical upper bound of the excess risk of the classifier introduced by the perturbation.…”

Section: Introductionmentioning

confidence: 99%

DPRF: A Differential Privacy Protection Random Forest

Hou

Meng

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Providing privacy protection for classification algorithms has become a research hotspot in current data mining. In this paper, differential privacy is applied to the random forest classification algorithm, and a random forest algorithm based on differential privacy is proposed to protect the privacy information in the data classification process. Firstly, differential privacy provides privacy protection by adding perturbation noise, which leads to a decrease in the classification accuracy of random forest algorithms. In order to reduce the impact of differential privacy protection on the accuracy of random forest classification, a hybrid decision tree algorithm is proposed. For the construction of a single decision tree in a random forest, the information gain ratio in the ID3 algorithm and the information gain ratio in the C4.5 are combined to generate a new attribute metric IG_GR to improve the classification accuracy of a single decision tree. Secondly, a new privacy budget allocation strategy is proposed. For nodes of different depths in the decision tree, the privacy budget is allocated to its counting query and attribute query by weight, which is used to balance the signalto-noise ratio of differential privacy technology to nodes of different depths in the decision tree. At the same time, the hybrid decision tree algorithm is applied to the construction of random forest, which balances the privacy and classification accuracy of the random forest algorithm based on differential privacy. Finally, this paper conducted experiments on UCI's Adult and Mushroom datasets. The results show that compared with the traditional decision tree algorithm, the algorithm proposed in this paper has better classification accuracy; and the DPRF can provide effective privacy protection under the premise of ensuring high classification performance. The work of this paper achieves a balance between privacy and classification accuracy, and has practical application value.

show abstract

Large Margin Multiclass Gaussian Classification with Differential Privacy

Cited by 6 publications

References 14 publications

Differentially private distributed logistic regression using private and public data

Differentially private distributed logistic regression using private and public data

An Interpretable Graph-based Mapping of Trustworthy Machine Learning Research

DPRF: A Differential Privacy Protection Random Forest

Contact Info

Product

Resources

About