對話系統論文集(4)-liubing-end-to-end

需要了解POMDP 問題: (b)end-to-end比 (a)不同成分分開訓練 的結果好。 a的時候上游組件的問題可能傳遞到下游的開始; Moreover, each component in the pipeline is ideally re-trained as preceding components are updated, so that we have inputs similar
相關文章
相關標籤/搜索