Neural Stored-program Memory
Hung Le, Truyen Tran, Svetha Venkatesh
Abstract
Neural networks powered with external memory simulate computer behaviors. These models, which use the memory to store data for a neural controller, can learn algorithms and other complex tasks. In this paper, we introduce a new memory to store weights for the controller, analogous to the stored-program memory in modern computer architectures. The proposed model, dubbed Neural Stored-program Memory, augments current memory-augmented neural networks, creating differentiable machines that can switch programs through time, adapt to variable contexts and thus fully resemble the Universal Turing Machine. A wide range of experiments demonstrate that the resulting machines not only excel in classical algorithmic problems, but also have potential for compositional, continual, few-shot learning and question-answering tasks.
BibTeX
@inproceedings{
Le2020Neural,
title={Neural Stored-program Memory},
author={Hung Le and Truyen Tran and Svetha Venkatesh},
booktitle={International Conference on Learning Representations},
year={2020},
url={https://openreview.net/forum?id=rkxxA24FDr}
}