Abstract
A mathematical model is presented for the description of parallel Boltzmann machines. The framework is based on the theory of Markov chains and combines a number of previously known results into one generic model. It is argued that parallel Boltzmann machines maximize a function consisting of a weighted sum of the well-known consensus function and a pseudoconsensus function. The weighting is determined by the amount of parallelism used in the Boltzmann machine and turns out to be problem dependent.
Original language | English |
---|---|
Pages (from-to) | 65-75 |
Number of pages | 11 |
Journal | Journal of Parallel and Distributed Computing |
Volume | 13 |
Issue number | 1 |
DOIs | |
Publication status | Published - 1991 |
Externally published | Yes |