Repositories
1M_Generalization is a simple anonymization algorithm for 1:M dataset. It contains two sub-algorithms: Mondrian (for relational part) and Partition (transaction part). Both of them are straight forward, and can be repalced by more powerful algorithm with limtied modification.
Anatomize and Partition Anonymization
Algorithms Princeton exercises
This repository is an open source python implement for Anatomy. I implement this algorithm in python for further study.
This repository is an python implement of Apriori_based_Anonymization for set-valued dataset anonymization.
The raw mondrian is designed for numerical attributes. When comes to categorical attributes, Mondrian needs to transform categorical attributes to numerical ones. This transformations is not good for some applications. In 2006, LeFevre proposed basic Mondrian, which support both categorical and numerical attributes. This repository is an implementation for basic Mondrian.
Core HW bindings and optimizations for BigDL
CS234: Reinforcement Learning Winter 2019
cluster based generalization for k-anonymity
Deep Learning Exercise and Notebook
Enhanced_Mondrian for incomplete microdata
An Industrial Grade Federated Learning Framework
My solution for Google APAC
Support Deep Learning on Hadoop platform
HILB and iDIST are two efficient anonymization algorithms proposed by Gabriel Ghinita in his paper. This repository is a python implementation for HILB and iDIST.
本文档适合于刚入学的硕士和博士(计算机专业最好，其他专业可参考)。
Supercharge Your LLM with the Fastest KV Cache Layer