Crossing the Validation Crisis: Cross-Validation Reduces Benchmarking Variance Surprisingly Well
Modern machine learning progresses through empirical work, benchmarking new methods to evaluate relative performance. However, the statistical variability inherent to evaluation - exacerbated by the stochastic nature of many algorithms - often makes performance estimation unreliable due to the limited test samples available, leading to a validation crisis in which genuine advances are difficult to discern. In this work, we show that cross-validation improves markedly confidence when evaluating and comparing learning algorithm performances. We introduce the concept of sample gain, which quantifies the virtual data augmentation achieved by using multiple cross-validation splits to reduce benchmarking variance. Experiments on both synthetic and real-world datasets (histopathologic scans and NLP fine-tuning) demonstrate that multiple splits can substantially improve the reliability and stability of performance estimates, with diminishing returns often setting in later than expected. We also introduce a procedure to dynamically early-stop cross-validation by estimating from the first few folds if subsequent folds will bring large sample gains. Our findings highlight the value of pushing cross-validation on available samples to achieve robust and reliable benchmarking.
Code (0)
등록된 구현이 없습니다.
Tasks
Data AugmentationSimilar Papers 제목 키워드 기반
What is Reproducibility in Artificial Intelligence and Machine Learning Research?
In the rapidly evolving fields of Artificial Intelligence (AI) and Machine Learning (ML), the reproducibility crisis underscores the urgent need for clear validation methodologies to maintain scientific integrity and enc…
Enumerating the k-fold configurations in multi-class classification problems
K-fold cross-validation is a widely used tool for assessing classifier performance. The reproducibility crisis faced by artificial intelligence partly results from the irreproducibility of reported k-fold cross-validatio…
Binary ClassificationMulti-class ClassificationA Vision-and-Knowledge Enhanced Large Language Model for Generalizable Pedestrian Crossing Behavior Inference
Existing paradigms for inferring pedestrian crossing behavior, ranging from statistical models to supervised learning methods, demonstrate limited generalizability and perform inadequately on new sites. Recent advances i…
Few-Shot LearningNo crisis should go to waste
The crisis in the reproducibility of experiments invites a re-evaluation of methods of inquiry and validation procedures. The text challenges current assumptions of knowledge acquisition and introduces G-complexity for d…
ExplainGAN: Model Explanation via Decision Boundary Crossing Transformations
We introduce a new method for interpreting computer vision models: visually perceptible, decision-boundary crossing transformations. Our goal is to answer a simple question: why did a model classify an image as being of …
model