Unreasonable effectiveness of learning neural networks: From accessible states and robust ensembles to basic algorithmic schemes

In artificial neural networks, learning from data is a computationally demanding task in which a large number of connection weights are iteratively tuned through stochastic-gradient-based heuristic processes over a cost function. It is not well understood how learning occurs in these systems, in par...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Proceedings of the National Academy of Sciences - PNAS 2016-11, Vol.113 (48), p.E7655-E7662
Hauptverfasser:	Baldassi, Carlo, Borgs, Christian, Chayes, Jennifer T., Ingrosso, Alessandro, Lucibello, Carlo, Saglietti, Luca, Zecchina, Riccardo
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Heuristic Markov analysis Markov chains Neural networks Optimization algorithms Physical Sciences PNAS Plus
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!