Search Results for author: Pouria Saidi

Found 2 papers, 0 papers with code

Unraveling overoptimism and publication bias in ML-driven science

no code implementations23 May 2024 Pouria Saidi, Gautam Dasarathy, Visar Berisha

Validity concerns are underscored by findings of an inverse relationship between sample size and reported accuracy in published ML models, contrasting with the theory of learning curves where accuracy should improve or remain stable with increasing sample size.

Active Sequential Two-Sample Testing

no code implementations30 Jan 2023 Weizhi Li, Prad Kadambi, Pouria Saidi, Karthikeyan Natesan Ramamurthy, Gautam Dasarathy, Visar Berisha

The classification model is adaptively updated and used to predict where the (unlabelled) features have a high dependency on labels; labeling the ``high-dependency'' features leads to the increased power of the proposed testing framework.

Two-sample testing valid +1

Cannot find the paper you are looking for? You can Submit a new open access paper.