Paper Info
Title
Chunking Mechanisms For A Self Improving Associative Memory Model
Abstract
Memory is a key component of any intelligent agent. The efficiency of an agent's memory determines its ability to learn and recall new knowledge. The field of AI has been awash with new innovations aimed at improving the memory and recall process of an agent. To this end, none of these innovations have yielded an agent whose memory and recall process is as efficient as that of humans. Scientists continue to study the brain with an aim of establishing what makes it so effective. In this study, we create a self-improving associative memory model that is designed to mimic the human memory and recall process. Our initial tests have indicated that the model can mimic chunking of memory items of the brain through use of linear algebra. Chunking is believed to be a key process in memory formation and recall in the brain. It involves breaking down memory items into granular elements that can be recombined in different ways thereby creating associations.
Year
Venue
Keywords
2017
2017 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (SSCI)
associative memory, cognitive memory, SVD, chunking mechanisms, merging mechanisms
Field
DocType
Citations 
Human memory,Intelligent agent,Content-addressable memory,Computer science,Natural language processing,Artificial intelligence,Chunking (psychology),Recall
Conference
0
PageRank 
References 
Authors
0.34
0
2
Name
Order
Citations
PageRank
Peter Kimani Mungai111.64
Runhe Huang240756.46