A Continual Learning Algorithm Based on Orthogonal Gradient Descent Beyond Neural Tangent Kernel Regime

Continual learning aims to enable neural networks to learn new tasks without catastrophic forgetting of previously learned knowledge. Orthogonal Gradient Descent algorithms have been proposed as an effective solution to mitigate catastrophic forgetting. However, these algorithms often rely on the Ne...

Ful tanımlama

Kaydedildi:

Detaylı Bibliyografya
Asıl Yazarlar:	Da Eun Lee, Kensuke Nakamura, Jae-Ho Tak, Byung-Woo Hong
Materyal Türü:	Artigo
Dil:	Inglês
Baskı/Yayın Bilgisi:	IEEE 2023-01-01
Seri Bilgileri:	IEEE Access
Konular:	Catastrophic forgetting continual learning neural tangent kernel orthogonal gradient descent orthogonal projection
Online Erişim:	https://ieeexplore.ieee.org/document/10213447/
Etiketler:	Etiketle Etiket eklenmemiş, İlk siz ekleyin!

A Continual Learning Algorithm Based on Orthogonal Gradient Descent Beyond Neural Tangent Kernel Regime

Benzer Materyaller