Year: 2023
Type: preprint
Abstract: Neural networks are notorious for being overconfident predictors, posing a significant challenge to their safe deployment in real-world applications. While feature normalization has garnered considerable attention within the deep learning literature, current train-time regularization methods for Out-of-Distribution(OOD) detection are yet to fully exploit this potential. Indeed, the naive incorporation of feature normalization within neural networks does not guarantee substantial improvement in OOD detection performance. In this work, we introduce T2FNorm, a novel approach to transforming features to hyperspherical space during training, while employing non-transformed space for OOD-scoring purposes. This method yields a surprising enhancement in OOD detection capabilities without compromising model accuracy in in-distribution(ID). Our investigation demonstrates that the proposed technique substantially diminishes the norm of the features of all samples, more so in the case of out-of-distribution samples, thereby addressing the prevalent concern of overconfidence in neural networks. The proposed method also significantly improves various post-hoc OOD detection methods.
Source: arXiv (Cornell University)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G1051717971, Robotic Assisted Imaging, Multispectral polarization-resolved endoscopy and vision for intraoperative imaging of tissue microstructure and function, Ultra-low-cost endoscopy for gastric cancer screening in rural China, Wellcome Trust Centre for Surgical and Interventional Sciences, G7705462073, G868016207, G8925563342, G985999274
OpenAlex ID: W4378768569