Related Experiment Video
Updated: Jul 16, 2026

The Double-H Maze: A Robust Behavioral Test for Learning and Memory in Rodents
Published on: July 8, 2015
Capacity analysis for a two-level decoupled Hamming network for associative memory under a noisy environment.
Liang Chen1, Naoyuki Tokuda, Akira Nagai
1Computer Science Department, University of Northern British Columbia, BC, Canada V2N 4Z9. lchen@ieee.org
The two-level decoupled Hamming network offers superior capacity and efficiency compared to single-level models. This advanced associative memory is ideal for hardware implementation due to its resilience against concentrated noise.
Area of Science:
- Artificial Intelligence
- Computer Science
- Neuroscience
Background:
- Single-level Hamming networks are foundational associative memory models.
- These networks face limitations in capacity and noise resilience.
- Previous research explored uniform random noise effects on Hamming networks.
Purpose of the Study:
- To evaluate the capacity of a two-level decoupled Hamming network.
- To compare its performance against single-level Hamming associative memory.
- To assess its robustness against different noise types.
Main Methods:
- Detailed computational analysis of network performance.
- Comparative evaluation under uniform random noise and concentrated noise conditions.
- Assessment of hardware implementation feasibility and computational efficiency.
Main Results:
- The two-level decoupled Hamming network demonstrates significantly higher capacity than single-level models.
- It exhibits greater resilience to concentrated noise, a prevalent issue.
- The network maintains computational efficiency and ease of hardware implementation.
Conclusions:
- The two-level decoupled Hamming network is a superior associative memory model.
- Its enhanced capacity and noise tolerance make it suitable for practical applications.
- Middle-sized windows are recommended for optimal performance in this network architecture.
Related Concept Videos
Interference and Decay
Interference occurs when competing memories hinder the retrieval of particular information. It can be classified into two types: proactive and retroactive interference. Proactive...
Understanding Memory
Network Function of a Circuit
Higher Mental Functions of Brain: Learning and Memory
Chunking and Rehearsal in Sensory Memory
Working Memory