方法证据记录
Mixture of Experts
Mixture of Experts (MoE) is a sparse neural-network architecture, introduced by Shazeer and colleagues in 2017 with the sparsely-gated MoE layer, in which only a subset of expert sub-networks is activated for each input. As seen in models such as Switch Transformer and Mixtral, it holds computation cost fixed even as the total parameter count grows.
源记录
引文逐字复制自方法源记录。这些引文不代表任何层级的验证。
Sparsely-Gated Mixture of Experts (MoE)
分类方法记录 · ml-model / deep-learning
- Shazeer, N. et al. (2017). Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer. ICLR. arXiv:1701.06538 · URL
- Jiang, A.Q. et al. (2024). Mixtral of Experts. arXiv. · URL
精选声明
声明已持久化到证据分类账中，每个声明都有自己的评估。
尚无精选声明
当分类账中没有声明时，此视图不会自行创建声明评估。
相关方法
从方法图中生成，显示为机器建议的关系 — 不推断任何证据声明。