%0 Journal Article %1 neyshabur2014search %A Neyshabur, Behnam %A Tomioka, Ryota %A Srebro, Nathan %D 2014 %K deep-learning foundations machine-learning stable theory regularisation %T In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning %U http://arxiv.org/abs/1412.6614 %X We present experiments demonstrating that some other form of capacity control, different from network size, plays a central role in learning multilayer feed-forward networks. We argue, partially through analogy to matrix factorization, that this is an inductive bias that can help shed light on deep learning.

@article{neyshabur2014search, abstract = {We present experiments demonstrating that some other form of capacity control, different from network size, plays a central role in learning multilayer feed-forward networks. We argue, partially through analogy to matrix factorization, that this is an inductive bias that can help shed light on deep learning.}, added-at = {2019-06-10T11:26:58.000+0200}, author = {Neyshabur, Behnam and Tomioka, Ryota and Srebro, Nathan}, biburl = {https://www.bibsonomy.org/bibtex/2caa79cc374936649b23d212f1437aadd/kirk86}, description = {[1412.6614] In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning}, interhash = {e302965c2d22ac1a6924d47fa8b61114}, intrahash = {caa79cc374936649b23d212f1437aadd}, keywords = {deep-learning foundations machine-learning stable theory regularisation}, note = {cite arxiv:1412.6614Comment: 9 pages, 2 figures}, timestamp = {2019-09-26T16:00:39.000+0200}, title = {In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning}, url = {http://arxiv.org/abs/1412.6614}, year = 2014 }