Unverified50% confidenceFactExact time
Model distillation is a technique in deep learning for transferring knowledge from a teacher model to a student model.
1
Sources
50%
Confidence
Long-term
Relevance
9/23/2026
First Seen
Sources
Related Entities
Related Claims
Verified知识蒸馏由Hinton等人在2015年提出，核心思想是让小型学生模型模仿大型教师模型的输出分布76% similarUnverified教材知识编排应先学传统机器学习再学深度学习，因为传统ML帮助建立特征工程、模型评估、偏差-方差权衡等核心概念直觉74% similarVerifiedKnowledge distillation, proposed by Hinton et al., has student models inherit implicit knowledge by learning from the soft probability distributions output by teacher models.73% similarUnverifiedThe teacher model in distillation does not have to be larger than the student model; it can be a domain expert model of comparable size.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/943501API
curl https://kongchang.com/api/v1/knowledge/claims/943501MCP
get_claim(id=943501)