[{"data":1,"prerenderedAt":37},["ShallowReactive",2],{"post-document:\u002Fdeep_learning\u002Ftransformer\u002F2026\u002F08\u002F06\u002Fmixture-of-experts\u002F":3},{"id":4,"title":5,"body":6,"categories":17,"date":20,"description":21,"extension":22,"image":23,"key_concepts":23,"last_modified_at":23,"legacyPath":24,"meta":25,"navigation":27,"part":23,"path":28,"published":27,"robots":23,"seo":29,"series":23,"stem":30,"strengths":23,"summary":31,"tags":32,"tradeoffs":23,"__hash__":36},"posts\u002Fposts\u002FDeep_Learning\u002FTransformer\u002F2026-08-06-mixture-of-experts.md","Mixture of Experts(MoE) 핵심 개념",{"type":7,"value":8,"toc":13},"minimark",[9],[10,11,12],"p",{},"Dense Model과 Sparse Model\nMixture of Experts의 기본 구조\nRouter는 무엇을 하는가\nTop-k Expert Routing\nLoad Balancing\nExpert Collapse\nMoE의 장점과 단점\nVision MoE\nEdge 환경에서의 MoE\nMoE 모델의 분산 학습",{"title":14,"searchDepth":15,"depth":15,"links":16},"",2,[],[18,19],"Deep_Learning","Transformer","2026-08-06 00:00:00 +0900","Sparse Model, Router, Top-k Routing, Load Balancing 등 Mixture of Experts의 핵심 개념을 정리한다.","md",null,"\u002Fdeep_learning\u002Ftransformer\u002F2026\u002F08\u002F06\u002Fmixture-of-experts\u002F",{"layout":26},"post",true,"\u002Fposts\u002Fdeep_learning\u002Ftransformer\u002F2026-08-06-mixture-of-experts",{"title":5,"description":21},"posts\u002FDeep_Learning\u002FTransformer\u002F2026-08-06-mixture-of-experts","MoE 모델의 구조와 라우팅 방식, 장단점, 분산 학습 개념을 정리한다.",[33,19,34,35],"Deep Learning","Mixture of Experts","MoE","bsfV7S2F11HVBqKYbrLPLCsZAdLbj4lFov3hMUsmUZY",1788744789215]