To Exit or Not to Exit: Cost-Effective Early-Exit Architecture Based on Markov Decision Process

Recently, studies on early-exit mechanisms have emerged to reduce the computational cost during the inference process of deep learning models. However, most existing early-exit architectures simply determine early exiting based only on a target confidence level in the prediction, without any conside...

Celý popis

Podrobná bibliografie

Hlavní autoři:	Kyu-Sik Kim, Hyun-Suk Lee
Médium:	Artigo
Jazyk:	Inglês
Vydáno:	MDPI AG 2024-07-01
Edice:	Mathematics
Témata:	deep learning early exit Markov decision process reinforcement learning
On-line přístup:	https://www.mdpi.com/2227-7390/12/14/2263
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!