Goodman, Rodney M.Rodney M.GoodmanTZI-DAR CHIUEH2020-06-162020-06-16199110459227https://scholars.lib.ntu.edu.tw/handle/123456789/502386https://www.scopus.com/inward/record.uri?eid=2-s2.0-0026122984&doi=10.1109%2f72.80338&partnerID=40&md5=435089bb8baae80ffd58db8751232726This paper presents a model for a class of hish-capacity associative memories. Since they are based on two-layer recurrent neural networks and their operations depend on the correlation measure, we call these associative memories recurrent correlation associative memories (RCAM’s). The RCAM’s are shown to be asymptotically stable in both synchronous and asynchronous (sequential) update modes as long as their weighting functions are continuous and monotone nondecreasing. In particular, a new high-capacity RCAM named the exponential correlation associative memory (ECAM) is proposed. The asymptotic storage capacity of the ECAM scales exponentially with the length of memory patterns, and it meets the ultimate upper bound for the capacity of associative memories. Furthermore, the asymptotic storage capacity of the ECAM with limited dynamic range in its exponentiation nodes is found to be proportional to that dynamic range. This paper also reports a 3 μm CMOS ECAM chip, which has been designed and fabricated. The prototype chip can store 32 24-bit memory patterns, and its speed is faster than one associative recall operation every 3 μs. An application of the ECAM chip to vector quantization is also described. © 1991 IEEE[SDGs]SDG7Computer Simulation; Data Storage, Digital - Associative; Information Theory; Integrated Circuits, CMOS; Probability; Exponential Correlation Associative Memories; Recurrent Correlation Associative Memories; Vector Quantization; Neural Networksjournal article10.1109/72.80338https://doi.org/10.1109/72.80338