Kickback Cuts Backprop's Red-Tape: Biologically Plausible Credit Assignment in Neural Networks. Balduzzi, D., Vanchinathan, H., & Buhmann, J. abstract bibtex Error backpropagation is an extremely effective algorithm for assigning credit in artificial neural networks. However, weight updates under Backprop depend on lengthy recursive computations and require separate output and error messages – features not shared by biological neurons, that are perhaps unnecessary. In this paper, we revisit Backprop and the credit assignment problem.
@article{balduzzi_kickback_nodate,
title = {Kickback {Cuts} {Backprop}'s {Red}-{Tape}: {Biologically} {Plausible} {Credit} {Assignment} in {Neural} {Networks}},
abstract = {Error backpropagation is an extremely effective algorithm for assigning credit in artificial neural networks. However, weight updates under Backprop depend on lengthy recursive computations and require separate output and error messages – features not shared by biological neurons, that are perhaps unnecessary. In this paper, we revisit Backprop and the credit assignment problem.},
language = {en},
author = {Balduzzi, David and Vanchinathan, Hastagiri and Buhmann, Joachim},
pages = {7}
}
Downloads: 0
{"_id":"8y5y29wnkkf2BW8hL","bibbaseid":"balduzzi-vanchinathan-buhmann-kickbackcutsbackpropsredtapebiologicallyplausiblecreditassignmentinneuralnetworks","authorIDs":[],"author_short":["Balduzzi, D.","Vanchinathan, H.","Buhmann, J."],"bibdata":{"bibtype":"article","type":"article","title":"Kickback Cuts Backprop's Red-Tape: Biologically Plausible Credit Assignment in Neural Networks","abstract":"Error backpropagation is an extremely effective algorithm for assigning credit in artificial neural networks. However, weight updates under Backprop depend on lengthy recursive computations and require separate output and error messages – features not shared by biological neurons, that are perhaps unnecessary. In this paper, we revisit Backprop and the credit assignment problem.","language":"en","author":[{"propositions":[],"lastnames":["Balduzzi"],"firstnames":["David"],"suffixes":[]},{"propositions":[],"lastnames":["Vanchinathan"],"firstnames":["Hastagiri"],"suffixes":[]},{"propositions":[],"lastnames":["Buhmann"],"firstnames":["Joachim"],"suffixes":[]}],"pages":"7","bibtex":"@article{balduzzi_kickback_nodate,\n\ttitle = {Kickback {Cuts} {Backprop}'s {Red}-{Tape}: {Biologically} {Plausible} {Credit} {Assignment} in {Neural} {Networks}},\n\tabstract = {Error backpropagation is an extremely effective algorithm for assigning credit in artificial neural networks. However, weight updates under Backprop depend on lengthy recursive computations and require separate output and error messages – features not shared by biological neurons, that are perhaps unnecessary. In this paper, we revisit Backprop and the credit assignment problem.},\n\tlanguage = {en},\n\tauthor = {Balduzzi, David and Vanchinathan, Hastagiri and Buhmann, Joachim},\n\tpages = {7}\n}\n\n","author_short":["Balduzzi, D.","Vanchinathan, H.","Buhmann, J."],"key":"balduzzi_kickback_nodate","id":"balduzzi_kickback_nodate","bibbaseid":"balduzzi-vanchinathan-buhmann-kickbackcutsbackpropsredtapebiologicallyplausiblecreditassignmentinneuralnetworks","role":"author","urls":{},"downloads":0,"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/asneha213","creationDate":"2019-07-08T00:48:54.938Z","downloads":0,"keywords":[],"search_terms":["kickback","cuts","backprop","red","tape","biologically","plausible","credit","assignment","neural","networks","balduzzi","vanchinathan","buhmann"],"title":"Kickback Cuts Backprop's Red-Tape: Biologically Plausible Credit Assignment in Neural Networks","year":null,"dataSources":["fjacg9txEnNSDwee6"]}