Learn on Source, Refine on Target:A Model Transfer Learning Framework with Random Forests

Abstract

We propose novel model transfer-learning methods that refine a decision forest model M learned within a "source" domain using a training set sampled from a "target" domain, assumed to be a variation of the source. We present two random forest transfer algorithms. The first algorithm searches greedily for locally optimal modifications of each tree structure by trying to locally expand or reduce the tree around individual nodes. The second algorithm does not modify structure, but only the parameter (thresholds) associated with decision nodes. We also propose to combine both methods by considering an ensemble that contains the union of the two forests. The proposed methods exhibit impressive experimental results over a range of problems.

2 columns, 14 pages, TPAMI submitted

References in corpus (6)

• Frustratingly Easy Domain Adaptation
• A Model of Inductive Bias Learning
• Risk Bounds for the Majority Vote: From a PAC-Bayesian Analysis to a Learning Algorithm
• Recognizing Activities and Spatial Context Using Wearable Sensors
• Learning from Multiple Outlooks
• Selective Transfer Learning for Cross Domain Recommendation

Cited by in corpus (7)

• Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges
• JobPruner: A Machine Learning Assistant for Exploring Parameter Spaces in HPC Applications
• Robust Kernel Density Estimation with Median-of-Means principle
• Transfer Learning-Based Outdoor Position Recovery with Telco Data
• Pocket Diagnosis: Secure Federated Learning against Poisoning Attack in the Cloud
• CHEER: Rich Model Helps Poor Model via Knowledge Infusion
• Adapted tree boosting for Transfer Learning