# ARMIN: Towards a More Efficient and Light-weight Recurrent Memory Network

## TL;DR

ARMIN introduces a lightweight, efficient memory-augmented neural network that simplifies memory addressing and improves integration, achieving comparable performance to LSTM with lower computational costs.

## Contribution

The paper proposes ARMIN, a novel MANN with automatic memory addressing and a new RNN cell, reducing complexity and computational overhead.

## Key findings

- ARMIN is more lightweight and efficient than existing memory networks.
- ARMIN achieves similar performance to LSTM with lower computational costs.
- Empirical results validate ARMIN's effectiveness across various tasks.

## Abstract

In recent years, memory-augmented neural networks(MANNs) have shown promising power to enhance the memory ability of neural networks for sequential processing tasks. However, previous MANNs suffer from complex memory addressing mechanism, making them relatively hard to train and causing computational overheads. Moreover, many of them reuse the classical RNN structure such as LSTM for memory processing, causing inefficient exploitations of memory information. In this paper, we introduce a novel MANN, the Auto-addressing and Recurrent Memory Integrating Network (ARMIN) to address these issues. The ARMIN only utilizes hidden state ht for automatic memory addressing, and uses a novel RNN cell for refined integration of memory information. Empirical results on a variety of experiments demonstrate that the ARMIN is more light-weight and efficient compared to existing memory networks. Moreover, we demonstrate that the ARMIN can achieve much lower computational overhead than vanilla LSTM while keeping similar performances. Codes are available on github.com/zoharli/armin.