tensorflow - Loss function (and encoding?) for angles - Cross Validated

I'm training a network to predict the angle of arrival of a signal. Labels are single values in the [-180, 180) interval. I'm seeing a discontinuity in predictions around ±180 degrees, which makes...

tensorflow - Loss function (and encoding?) for angles - Cross Validated Stack Internal Knowledge at work Bring the best of human thought and AI automation together at your work. Explore Stack Internal Loss function (and encoding?) for angles Ask Question Asked 6 years, 10 months ago Modified 5 years, 5 months ago Viewed 7k times 11 $\begingroup$ I'm training a network to predict the angle of arrival of a signal. Labels are single values in the [-180, 180) interval. I'm seeing a discontinuity in predictions around ±180 degrees, which makes sense as losses around that gap are incorrectly calcula

related reading

• What is the difference between a loss function and an error function? - Cross Validatedstats.stackexchange.com
• The Little Book of Deep Learningfleuret.org
• Untangling the Moons: A Visual History of Contrastive Learningtahabouhsine.com
• Some Math behind Neural Tangent Kernel | Lil'Loglilianweng.github.io
• Learning the integral of a diffusion model – Sander Dielemansander.ai
• 1412.0233arxiv.org
• CS231n Deep Learning for Computer Visioncs231n.github.io
• The Annotated Transformernlp.seas.harvard.edu
• Tinkerthinkingmachines.ai
• Sequence Modeling with CTCdistill.pub
• Understanding the Neural Tangent Kernel – EigenTaleseigentales.com
• [2605.01172] A Theory of Generalization in Deep Learningarxiv.org