Long Short-Term Memory Over Recursive Structures

The chain-structured long short-term memory (LSTM) has showed to be effective in a wide range of problems such as speech recognition and machine translation. In this paper, we pro-pose to extend it to tree structures, in which a memory cell can reflect the history memories of multiple child cells or multiple descendant cells in a recursive process. We call the model S-LSTM, which provides a principled way of considering long-distance interaction over hier-archies, e.g., language or image parse structures. We leverage the models for semantic composi-tion to understand the meaning of text, a funda-mental problem in natural language understand-ing, and show that it outperforms a state-of-the-art recursive model by replacing its composition layers with the S-LSTM memory blocks. We also show that utilizing the given structures is helpful in achieving a performance better than that with-out considering the structures. 1.