Robustness of Differentiable Neural Computer Using Limited Retention Vector-based Memory Deallocation in Language Model

Recurrent neural network (RNN) architectures have been used for language modeling (LM) tasks that require learning long-range word or character sequences. However, the RNN architecture is still suffered from unstable gradients on long-range sequences. To address the issue of long-range sequences, an...

Full description

Saved in:
Bibliographic Details
Published in:KSII transactions on Internet and information systems 2021-03, Vol.15 (3), p.837-852
Main Authors: Lee, Donghyun, Park, Hosung, Seo, Soonshin, Son, Hyunsoo, Kim, Gyujin, Kim, Ji-Hwan
Format: Article
Language:Korean
Subjects:
Differentiable Neural Computer
DNC
Language Model
LM
Memory Deallocation
Retention Vector
Robustness
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recurrent neural network (RNN) architectures have been used for language modeling (LM) tasks that require learning long-range word or character sequences. However, the RNN architecture is still suffered from unstable gradients on long-range sequences. To address the issue of long-range sequences, an attention mechanism has been used, showing state-of-the-art (SOTA) performance in all LM tasks. A differentiable neural computer (DNC) is a deep learning architecture using an attention mechanism. The DNC architecture is a neural network augmented with a content-addressable external memory. However, in the write operation, some information unrelated to the input word remains in memory. Moreover, DNCs have been found to perform poorly with low numbers of weight parameters. Therefore, we propose a robust memory deallocation method using a limited retention vector. The limited retention vector determines whether the network increases or decreases its usage of information in external memory according to a threshold. We experimentally evaluate the robustness of a DNC implementing the proposed approach according to the size of the controller and external memory on the enwik8 LM task. When we decreased the number of weight parameters by 32.47%, the proposed DNC showed a low bits-per-character (BPC) degradation of 4.30%, demonstrating the effectiveness of our approach in language modeling tasks.
ISSN:1976-7277
1976-7277