Simple Classifier
Simple classifiers, aiming for accurate and efficient classification with minimal computational complexity, are a cornerstone of machine learning. Current research emphasizes improving their robustness to noisy data, imbalanced datasets, and distribution shifts, often employing techniques like data augmentation, ensemble methods (e.g., Random Forests, Gradient Boosting), and logistic regression coupled with embeddings from smaller language models. These advancements are crucial for deploying reliable classifiers in resource-constrained environments and for enhancing the interpretability and trustworthiness of AI systems across diverse applications, from medical diagnosis to fraud detection.
Papers
The Impact of an XAI-Augmented Approach on Binary Classification with Scarce Data
Ximing Wen, Rosina O. Weber, Anik Sen, Darryl Hannan, Steven C. Nesbit, Vincent Chan, Alberto Goffi, Michael Morris, John C. Hunninghake, Nicholas E. Villalobos, Edward Kim, Christopher J. MacLellan
Protecting Privacy in Classifiers by Token Manipulation
Re'em Harel, Yair Elboher, Yuval Pinter
Swin transformers are robust to distribution and concept drift in endoscopy-based longitudinal rectal cancer assessment
Jorge Tapias Gomez, Aneesh Rangnekar, Hannah Williams, Hannah Thompson, Julio Garcia-Aguilar, Joshua Jesse Smith, Harini Veeraraghavan
Tilt your Head: Activating the Hidden Spatial-Invariance of Classifiers
Johann Schmidt, Sebastian Stober