Analysis on Bidirectional Associative Memories with Multiplicative Weight Noise

Abstract. In neural networks, network faults can be exhibited in different forms, such as node fault and weight fault. One kind of weight faults is due to the hardware or software precision. This kind of weight faults can be modelled as multiplicative weight noise. This paper analyzes the capacity of a bidirectional associative memory (BAM) affected by multiplicative weight noise. Assuming that weights are corrupted by multiplicative noise, we study how many number of pattern pairs can be stored as fixed points. Since capacity is not meaningful without considering the error correction capability, we also present the capacity of a BAM with multiplicative noise when there are some errors in the input pattern. Simulation results have been carried out to confirm our derivations.

• Type: Conference
• Year: 2007
• Where: ICONIP
• Authors: Chi-Sing Leung, Pui-Fai Sum, Tien-Tsin Wong
• Keywords: ICONIP 2007, Information Technology, Multiplicative Noise, Multiplicative Weight Noise, Weight Faults