AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Optimal Decision Tree with Noisy Outcomes

Su Jia, viswanath nagarajan, Fatemeh Navidi, R Ravi

NeurIPS 2019 · Learning theory and algorithms › Prediction and decision losses · approach: Adaptive testing and experiments

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 changedWhereMeasuredEffect
ODTNr (adaptive, O(r+log m)) instead of Non-AdapODT with noisy outcomes · WISER (unique stopping)expected number of tests: 11.568 -> 8.357Improved
ODTNr instead of Non-AdapODT with noisy outcomes · CL-30expected number of tests: 10.204 -> 8.000Improved
ODTNr instead of ODTNhODT with noisy outcomes · CL-30expected number of tests: 8.729 -> 8.000Improved
ODTNr instead of ODTNhODT with noisy outcomes · WISER (unique stopping)expected number of tests: 9.707 -> 8.357Improved
ODTNr vs ODTNh with no noise (CL-0)ODT with noisy outcomes · CL-0expected number of tests: 7.910 -> 7.910No clear effect
ODTNr instead of Non-Adap under neighborhood stoppingODT with noisy outcomes · Original WISER, neighborhood stoppingexpected number of tests: 16.995 -> 11.163Improved
ODTNr instead of Non-Adap under clique stoppingODT with noisy outcomes · Original WISER, clique stoppingexpected number of tests: 21.281 -> 11.817Improved
ODTNr instead of ODTNh under clique stoppingODT with noisy outcomes · Original WISER, clique stoppingexpected number of tests: 12.506 -> 11.817Improved
skip non-informative tests adaptively (Low-Adap vs Non-Adap)Non-Adap · WISER (unique stopping)expected number of tests: 11.568 -> 9.152Improved
Low-Adap vs Non-Adap with neighborhood stoppingNon-Adap · Original WISER, neighborhood stoppingexpected number of tests: 16.995 -> 16.983No clear effect
increase classifier margin noise threshold d from 0 to 30ODTNh · CL-dexpected number of tests: 7.910 -> 8.729Hurt
increase classifier margin noise threshold d from 0 to 30ODTNr · CL-dexpected number of tests: 7.910 -> 8.000No clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy