TY - JOUR
T1 - Learning from rounded-off data
AU - Cheung, Dennis
AU - Cucker, Felipe
PY - 2003/4/10
Y1 - 2003/4/10
N2 - We provide an algorithm to PAC learn multivariate polynomials with real coefficients. The instance space from which labeled samples are drawn is ℝV but the coordinates of such samples are known only approximately. The algorithm is iterative and the main ingredient of its complexity, the number of iterations it performs, is estimated using the condition number of a linear programming problem associated to the sample. To the best of our knowledge, this is the first study of PAC learning concepts parameterized by real numbers from approximate data.
AB - We provide an algorithm to PAC learn multivariate polynomials with real coefficients. The instance space from which labeled samples are drawn is ℝV but the coordinates of such samples are known only approximately. The algorithm is iterative and the main ingredient of its complexity, the number of iterations it performs, is estimated using the condition number of a linear programming problem associated to the sample. To the best of our knowledge, this is the first study of PAC learning concepts parameterized by real numbers from approximate data.
UR - http://www.scopus.com/inward/record.url?scp=0038681899&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-0038681899&origin=recordpage
U2 - 10.1016/S0890-5401(03)00015-4
DO - 10.1016/S0890-5401(03)00015-4
M3 - RGC 21 - Publication in refereed journal
SN - 0890-5401
VL - 182
SP - 1
EP - 13
JO - Information and Computation
JF - Information and Computation
IS - 1
ER -