Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Residual Likelihood Forests

Nov 04, 2020
Yan Zuo, Tom Drummond

Share this with someone who'll enjoy it:

This paper presents a novel ensemble learning approach called Residual Likelihood Forests (RLF). Our weak learners produce conditional likelihoods that are sequentially optimized using global loss in the context of previous learners within a boosting-like framework (rather than probability distributions that are measured from observed data) and are combined multiplicatively (rather than additively). This increases the efficiency of our strong classifier, allowing for the design of classifiers which are more compact in terms of model capacity. We apply our method to several machine learning classification tasks, showing significant improvements in performance. When compared against several ensemble approaches including Random Forests and Gradient Boosted Trees, RLFs offer a significant improvement in performance whilst concurrently reducing the required model size.

* Accepted at BMVC2020 

   Access Paper Source

Share this with someone who'll enjoy it: