Research map / Label-efficient and robust learning / Recognition and long-tail
Are Random Forests Truly the Best Classifiers?
Michael Wainberg, Babak Alipanahi, Brendan J. Frey
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| evaluate only on benchmarks run successfully by all 8 classifiers (ablation) | rf_caret · UCI 121 datasets | mean accuracy (%): 82.30 -> 81.78 | Hurt |
| evaluate only on benchmarks run successfully by all 8 classifiers (ablation) | parRF_caret · UCI 121 datasets | mean accuracy (%): 81.96 -> 81.32 | Hurt |
| evaluate only on benchmarks run successfully by all 8 classifiers (ablation) | elm_kernel_matlab · UCI 121 datasets | mean accuracy (%): 81.96 -> 81.84 | No clear effect |
| evaluate only on benchmarks run successfully by all 8 classifiers (ablation) | svm_C · UCI 121 datasets | mean accuracy (%): 81.81 -> 81.30 | Hurt |
| replace paired t-test with Wilcoxon signed-rank and drop errors instead of imputing 82% (ablation) | significance of parRF_caret vs others · UCI 121 datasets | p value: 0.834 -> 0.8 | No clear effect |
| parRF_caret vs svm_C mean accuracy difference across 119 benchmarks | parRF_caret · UCI benchmarks where both ran | accuracy difference (%): 220.4 vs 219.6, about 0.007 | No clear effect |
| compare rf_caret vs svm_C by sum of accuracy differences | rf_caret · UCI 121 datasets | sum of positive vs negative accuracy differences: 193.8 vs 139.8 | No clear effect |
| compare parRF_caret with C5.0_caret under any test | parRF_caret · UCI 121 datasets | p value: 0.001 -> 0.0001 | Improved |
| compare parRF_caret with avNNet_caret | parRF_caret · UCI 121 datasets | p value: 0.03 -> 0.02 | Improved |
Closest papers by meaning
- Comparison-Based Random Forests (ICML 2018 · similarity 0.80)
- Towards Convergence Rate Analysis of Random Forests for Classification (NeurIPS 2020 · similarity 0.80)
- Random Composite Forests (AAAI 2016 · similarity 0.80)
- Randomization as Regularization: A Degrees of Freedom Explanation for Random Forest Success (JMLR 2020 · similarity 0.80)
- When do random forests fail? (NeurIPS 2018 · similarity 0.79)
- Statistical Multicriteria Benchmarking via the GSD-Front (NeurIPS 2024 · similarity 0.76)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
