Private yet Efficient Decision Tree Evaluation

@inproceedings{Joye2018PrivateYE,
  title={Private yet Efficient Decision Tree Evaluation},
  author={Marc Joye and Fariborz Salehi},
  booktitle={DBSec},
  year={2018}
}
Decision trees are a popular method for a variety of machine learning tasks. [] Key Method Our design reduces the complexity of existing solutions with a more interactive setting, which improves the total number of comparisons to evaluate the decision tree. It crucially uses oblivious transfer protocols and leverages their amortized overhead. Furthermore, and of independent interest, we improve by roughly a factor of two the DGK comparison protocol.

Figures and Tables from this paper

Let's Stride Blindfolded in a Forest: Sublinear Multi-Client Decision Trees Evaluation
TLDR
A privacy-preserving decision-tree evaluation protocol purely based on additive homomorphic encryption, without introducing dummy nodes for hiding the tree structure, but it runs a secure comparison for each decision node, resulting in linear complexity.
SoK: Modular and Efficient Private Decision Tree Evaluation
TLDR
This work systematically review and analyse state-of-the-art protocols for the three phases of private decision tree evaluation protocols: feature selection, comparison, and path evaluation, and identifies novel combinations of these protocols that provide better tradeoffs than existing protocols.
Scalable Private Decision Tree Evaluation with Sublinear Communication
TLDR
This work designs a sublinear PDTE protocol based on the proposed SOS functionality, allowing two parties to perform a secret-shared oblivious read operation from an array and proves its security under a semi-honest adversary.
Non-Interactive Private Decision Tree Evaluation
TLDR
The goal is to design and implement a novel client-server protocol that delegates the complete tree evaluation to the server while preserving privacy and reducing the overhead, and is able to provide the first non-interactive protocol.
UC Secure Private Branching Program and Decision Tree Evaluation
TLDR
This work proposes a UC-secure efficient multi-party computation platform for outsourced branching program and/or decision tree evaluation, and constructs a constant- round protocol and a poly-round protocol.
Privacy-Preserving Decision Trees Training and Prediction
TLDR
This work focuses on the ever-popular tree based methods, and proposes a new privacy-preserving solution to training and prediction for trees, which employs a low-degree approximation for the step-function together with a lightweight interactive protocol, to replace components of the vanilla algorithm that are costly over encrypted data.
Secure and Fast Decision Tree Evaluation on Outsourced Cloud Data
TLDR
This work designs a privacy-preserving decision tree evaluation scheme that can fully protect the tree model and clients’ data privacy simultaneously, and also supports offline service users which is essential to the system’s scalability.
Privacy-Preserving Machine Learning
TLDR
This keynote focuses on how cryptography can enable privacy-preserving machine learning services, in particular, decision tree evaluation.
Privacy-Preserving Decision Tree Training and Prediction against Malicious Server
TLDR
This work presents new protocols for privacy-preserving decision trees, for both training and prediction, achieving the following desirable properties: efficiency, efficiency, security, and practical usability, which are the first to offer all these properties simultaneously.
P INFER : Privacy-Preserving Inference for Machine Learning ? Logistic Regression , Support Vector Machines , and More , Over Encrypted Data
TLDR
A variety of protocols for privacy-preserving regression and classification that only require additively homomorphic encryption algorithms, limit interactions to a mere request and response, and that can be used directly for important machine-learning algorithms such as logistic regression and SVM classification are proposed.
...
1
2
3
...

References

SHOWING 1-10 OF 37 REFERENCES
Privately Evaluating Decision Trees and Random Forests
TLDR
Two protocols for privately evaluating decision trees and random forests are developed and an extension of the semi-honest protocol is given that is robust against malicious adversaries and demonstrates a tenfold improvement in computation and bandwidth.
Privacy-Preserving Decision Trees Evaluation via Linear Functions
TLDR
This work proposes privacy-preserving decision tree evaluation protocols which hide the sensitive inputs (model and query) from the counterparty by cleverly exploiting the structure of decision trees, which avoids an exponential number of encryptions in the depth of the decision tree.
Building decision tree classifier on private data
TLDR
This paper presents a protocol that allows Alice and Bob to conduct such a classifier building without having to compromise their privacy, and is built upon a useful building block, the scalar product protocol.
The Random Subspace Method for Constructing Decision Forests
  • T. Ho
  • Computer Science
    IEEE Trans. Pattern Anal. Mach. Intell.
  • 1998
TLDR
A method to construct a decision tree based classifier is proposed that maintains highest accuracy on training data and improves on generalization accuracy as it grows in complexity.
SecureML: A System for Scalable Privacy-Preserving Machine Learning
TLDR
This paper presents new and efficient protocols for privacy preserving machine learning for linear regression, logistic regression and neural network training using the stochastic gradient descent method, and implements the first privacy preserving system for training neural networks.
Privacy-preserving matrix factorization
TLDR
This work shows that a recommender can profile items without ever learning the ratings users provide, or even which items they have rated, by designing a system that performs matrix factorization, a popular method used in a variety of modern recommendation systems, through a cryptographic technique known as garbled circuits.
Secure Evaluation of Private Linear Branching Programs with Medical Applications
TLDR
A new, more efficient privacy-protecting protocols for remote evaluation of classification/diagnostic programs, and a useful generalization of BP that is introduced.
Machine Learning Classification over Encrypted Data
TLDR
A new library of building blocks is constructed, which enables constructing a wide range of privacy-preserving classifiers and it is demonstrated how this library can be used to construct other classifiers than the three mentioned above, such as a multiplexer and a face detection classifier.
Privacy Preserving Data Mining
TLDR
This work considers a scenario in which two parties owning confidential databases wish to run a data mining algorithm on the union of their databases, without revealing any unnecessary information, and proposes a protocol that is considerably more efficient than generic solutions and demands both very few rounds of communication and reasonable bandwidth.
Privacy-preserving SVM classification
TLDR
This paper proposes a privacy-preserving solution for support vector machine (SVM) classification, PP-SVM, which constructs the global SVM classification model from data distributed at multiple parties, without disclosing the data of each party to others.
...
1
2
3
4
...