Learning to Write by Learning the Objective
international conference on learning representations, 2018.
Recurrent Neural Networks (RNNs) are powerful autoregressive sequence models for learning prevalent patterns in natural language. Yet language generated by RNNs often shows several degenerate characteristics that are uncommon in human language; while fluent, RNN language production can be overly generic, repetitive, and even self-contradi...More
PPT (Upload PPT)