4 Local differential privacy for machine learning (part-1)
This chapter covers:
- Introducing the concept and definition of Local Differential Privacy (LDP).
- Implementing the randomized response mechanism for LDP.
- Introducing the theory behind LDP mechanisms for one-dimensional data frequency estimation.
- Implementing and experimenting different LDP mechanisms for one-dimensional data with direct encoding, histogram encoding and unary encoding mechanisms.
In the previous chapter, we looked into centralized Differential Privacy, where there is a trusted data curator who collects data from individuals and applies different techniques to obtain differentially private statistics about the population. Then, the data curator publishes privacy-preserving statistics about this population. However, these techniques are not suitable when individuals do not trust the data curator completely. To eliminate the need of trusted data curator, various techniques to satisfy differential privacy in the local setting have been studied. In this chapter, we will walk through the concept, the mechanisms and the applications of the local setting of differential privacy, namely, local differential privacy.
