Securely Training Decision Trees Efficiently
- Divyanshu Bhardwaj ,
- Sandhya Saravanan ,
- Nishanth Chandran ,
- Divya Gupta
31st Annual Conference on Computer and Communications Security (ACM CCS 2024) |
Decision trees are an important class of supervised learning algorithms. When multiple entities contribute data to train a decision tree (e.g. for fraud detection in the financial sector), data privacy concerns necessitate the use of a privacy-enhancing technology such as secure multi-party computation (MPC) in order to secure the underlying training data. Prior state-of-the-art (Hamada et al.) construct an MPC protocol for decision tree training with a communication of O (βππ log π ), when building a decision tree of height β for a training dataset of π samples, each having π attributes.
In this work, we significantly reduce the communication complexity of secure decision tree training. We construct a protocol with communication complexity O (ππ log π + βππ + βπ log π ), thereby achieving an improvement of β min(β, π, log π ) over Hamada et al. At the core of our technique is an improved protocol to regroup sorted private elements further into additional groups (according to a flag vector) while maintaining their relative ordering. We implement our protocol in the MP-SPDZ framework and show that it requires 10Γ lesser communication and is 9Γ faster than Hamada et al.