The Over-Certainty Phenomenon in Modern UDA Algorithms

arXiv:2404.16168v1 Announce Type: new
Abstract: When neural networks are confronted with unfamiliar data that deviate from their training set, this signifies a domain shift. While these networks output predictions on their inputs, they typically fail to account for their level of familiarity with these novel observations. This challenge becomes even more pronounced in resource-constrained settings, such as embedded systems or edge devices. To address such challenges, we aim to recalibrate a neural network's decision boundaries in relation to its cognizance …

abstract algorithms arxiv challenge cs.ai cs.lg data domain inputs modern networks neural networks novel predictions set shift stat.ml training type while