TY - JOUR
T1 - Training a perceptron in a discrete weight space
AU - Rosen-Zvi, M.
AU - Kanter, I.
PY - 2001/10/1
Y1 - 2001/10/1
N2 - Learning in a perceptron in a discrete weight space is examined analytically and numerically. The learning algorithm is based on the training of the continuous perceptron and prediction following the clipped weights. The realtions between the overlaps of the continuous teacher with the discrete/continuous students were derived. The results show that learning in the case of finite depth is possible by using a continuous precursors.
AB - Learning in a perceptron in a discrete weight space is examined analytically and numerically. The learning algorithm is based on the training of the continuous perceptron and prediction following the clipped weights. The realtions between the overlaps of the continuous teacher with the discrete/continuous students were derived. The results show that learning in the case of finite depth is possible by using a continuous precursors.
UR - http://www.scopus.com/inward/record.url?scp=35475049&partnerID=8YFLogxK
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
VL - 64
JO - Physical Review E - Statistical, Nonlinear, and Soft Matter Physics
JF - Physical Review E - Statistical, Nonlinear, and Soft Matter Physics
IS - 4 II
ER -