TY - CHAP
T1 - How Neural Networks (NN) Can (Hopefully) Learn Faster by Taking into Account Known Constraints
AU - Baral, Chitta
AU - Ceberio, Martine
AU - Kreinovich, Vladik
N1 - Funding Information:
Acknowledgements This work was supported in part by NSF grants HRD-0734825, HRD-1242122, and DUE-0926721, and by an award from Prudential Foundation.
Publisher Copyright:
© 2020, Springer Nature Switzerland AG.
PY - 2020
Y1 - 2020
N2 - Neural networks are a very successful machine learning technique. At present, deep (multi-layer) neural networks are the most successful among the known machine learning techniques. However, they still have some limitations, One of their main limitations is that their learning process still too slow. The major reason why learning in neural networks is slow is that neural networks are currently unable to take prior knowledge into account. As a result, they simply ignore this knowledge and simulate learning “from scratch”. In this paper, we show how neural networks can take prior knowledge into account and thus, hopefully, learn faster.
AB - Neural networks are a very successful machine learning technique. At present, deep (multi-layer) neural networks are the most successful among the known machine learning techniques. However, they still have some limitations, One of their main limitations is that their learning process still too slow. The major reason why learning in neural networks is slow is that neural networks are currently unable to take prior knowledge into account. As a result, they simply ignore this knowledge and simulate learning “from scratch”. In this paper, we show how neural networks can take prior knowledge into account and thus, hopefully, learn faster.
UR - http://www.scopus.com/inward/record.url?scp=85085038099&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85085038099&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-40814-5_3
DO - 10.1007/978-3-030-40814-5_3
M3 - Chapter
AN - SCOPUS:85085038099
T3 - Studies in Systems, Decision and Control
SP - 15
EP - 20
BT - Studies in Systems, Decision and Control
PB - Springer
ER -