Dynamic Neural Turing Machine with Soft and Hard Addressing Schemes

Abstract

We extend neural Turing machine (NTM) model into a dynamic neural Turing machine (D-NTM) by introducing a trainable memory addressing scheme. This addressing scheme maintains for each memory cell two separate vectors, content and address vectors. This allows the D-NTM to learn a wide variety of location-based addressing strategies including both linear and nonlinear ones. We implement the D-NTM with both continuous, differentiable and discrete, non-differentiable read/write mechanisms. We investigate the mechanisms and effects of learning to read and write into a memory through experiments on Facebook bAbI tasks using both a feedforward and GRUcontroller. The D-NTM is evaluated on a set of Facebook bAbI tasks and shown to outperform NTM and LSTM baselines. We have done extensive analysis of our model and different variations of NTM on bAbI task. We also provide further experimental results on sequential pMNIST, Stanford Natural Language Inference, associative recall and copy tasks.

13 pages, 3 figures

References in corpus (3)

• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Large-scale Simple Question Answering with Memory Networks
• A Simple Way to Initialize Recurrent Networks of Rectified Linear Units

Cited by in corpus (5)

• End-to-End Answer Chunk Extraction and Ranking for Reading Comprehension
• Episodic Memory Reader: Learning What to Remember for Question Answering from Streaming Data
• Modeling Past and Future for Neural Machine Translation
• Beyond the Memory Wall: A Case for Memory-centric HPC System for Deep Learning
• Understanding Memory Modules on Learning Simple Algorithms