TY - GEN
T1 - Adaptive precision neural networks for image classification
AU - Gilberti, Michael
AU - Doboli, Alex
PY - 2008
Y1 - 2008
N2 - We present a technique and algorithms to solve the following problem: Given both a Neural Network trained to classify a set of images, along with a set of floating-point hardware blocks (in reconfigurable logic), find the arrangement of blocks that achieves the best mix of precision, resources and speed with respect to a given cost function. We first illustrate the technique in detail by using a small example, then show that it may be used for a larger problem, bar code classification.
AB - We present a technique and algorithms to solve the following problem: Given both a Neural Network trained to classify a set of images, along with a set of floating-point hardware blocks (in reconfigurable logic), find the arrangement of blocks that achieves the best mix of precision, resources and speed with respect to a given cost function. We first illustrate the technique in detail by using a small example, then show that it may be used for a larger problem, bar code classification.
UR - https://www.scopus.com/pages/publications/51949091235
U2 - 10.1109/AHS.2008.65
DO - 10.1109/AHS.2008.65
M3 - Conference contribution
AN - SCOPUS:51949091235
SN - 9780769531663
T3 - Proceedings of the 2008 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2008
SP - 244
EP - 251
BT - Proceedings of the 2008 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2008
T2 - 2008 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2008
Y2 - 22 June 2008 through 25 June 2008
ER -