Tackling Dataset Bias With an Automated Collection of Real-World Samples
Sign Up to like & getrecommendations! Published in 2022 at "IEEE Access"
DOI: 10.1109/access.2022.3226517
Abstract: The early 21st-century technological advancements tilted the scales towards data-driven learning. Thus, modern machine-learning systems rely heavily on data to learn complex models to efficiently provide relevant predictions. Data-driven learning suffers from overfitting, a situation… read more here.
Keywords: dataset bias; bias automated; tackling dataset; dataset ... See more keywords