כתבה
arXiv cs.AI ·
MemFLoRA: Memory-Floor LoRA לאדפטציה של CNN בקצה
MemFLoRA: Memory-Floor LoRA for CNN Adaptation at the Edge
MemFLoRA הוא פרקטיקה לאדפטציה של CNN בקצה, המציעה רידוק של 98.5-98.7% בזיכרון הפעילות.
תקציר מקורי באנגליתarXiv:2610.08669v1 Announce Type: cross Abstract: On-device learning is necessary when the model encounters user-,sensor-, or environment-specific shifts after deployment. Although parameter-efficient fine-tuning (PEFT) methods, particularly Low-Rank Adaptation (LoRA) variants, enable efficient adaptation at the edge, the limiting resource for Convolutional Neural Network (CNN) adaptation is often not the number of trainable parameters but the activation state that must be retained until the backward pass. This paper introduces Memory-Floor LoRA (MemFLoRA), a low-rank CNN adapter built around a memory-first design principle rather than a direct application of transformer-oriented LoRA. Instead of merely reducing trainable weights, we define an activation-memory-floor criterion: trainable b
קרא במקור המקורי
arxiv.org
פתח כתבה מקורית