Title :
Random clique codes
Author :
Gripon, Vincent ; Skachek, Vitaly ; Gross, Warren J. ; Rabbat, Michael
Author_Institution :
Dept. of Electr. & Comput. Eng., McGill Univ., Montréal, QC, Canada
Abstract :
A new family of associative memories based on sparse neural networks has been recently introduced. These memories achieve excellent performance thanks to the use of error-correcting coding principles. In this work, we introduce a new family of codes termed clique codes. These codes are based on the cliques in balanced n-partite graphs describing associative memories. In particular, we study an ensemble of random clique codes, and prove that such ensemble contains asymptotically good codes. Furthermore, these codes can be efficiently decoded using the neural networks based associative memories with limited complexity and memory consumption.
Keywords :
content-addressable storage; error correction codes; graph theory; neural nets; random codes; associative memories; balanced n-partite graphs; error correcting coding principle; random clique codes; sparse neural networks; Associative memory; Complexity theory; Hamming distance; Maximum likelihood decoding; Neural networks; Neurons;
Conference_Titel :
Turbo Codes and Iterative Information Processing (ISTC), 2012 7th International Symposium on
Conference_Location :
Gothenburg
Print_ISBN :
978-1-4577-2114-4
Electronic_ISBN :
2165-4700
DOI :
10.1109/ISTC.2012.6325211
