Programming Weights to Analog In-Memory Computing Cores by Direct Minimization of the Matrix-Vector Multiplication Error.
Julian BüchelAthanasios VasilopoulosBenedikt KerstingCorey LammieKevin BrewTimothy PhilipNicole SaulnierVijay NarayananManuel Le GalloAbu SebastianPublished in: IEEE J. Emerg. Sel. Topics Circuits Syst. (2023)
Keyphrases
- multicore processors
- computing power
- sparse matrix
- weight vector
- operating system
- weight matrix
- programming language
- error rate
- singular value decomposition
- address space
- matrix multiplication
- error function
- weighted sum
- objective function
- programming environment
- parallel algorithm
- matrix representation
- parallel architectures
- rows and columns
- signal processing
- moore penrose
- error minimization
- dot product
- symmetric matrix
- transformation matrix
- linear algebra
- object oriented programming
- covariance matrix
- vector space
- linear combination
- least squares