paper-with-me

홈 › Papers

Surprising properties of dropout in deep networks

2016-02-14 · David P. Helmbold, Philip M. Long

We analyze dropout in deep networks with rectified linear units and the quadratic loss. Our results expose surprising differences between the behavior of dropout and more traditional regularizers like weight decay. For example, on some simple data sets dropout training produces negative weights even though the output is the sum of the inputs. This provides a counterpoint to the suggestion that dropout discourages co-adaptation of weights. We also show that the dropout penalty can grow exponentially in the depth of the network while the weight-decay penalty remains essentially linear, and that dropout is insensitive to various re-scalings of the input features, outputs, and network weights. This last insensitivity implies that there are no isolated local minima of the dropout training criterion. Our work uncovers new properties of dropout, extends our understanding of why dropout succeeds, and lays the foundation for further progress.

📄 PDF Abstract BibTeX arXiv:1602.04484

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

To Drop or Not to Drop: Robustness, Consistency and Differential Privacy Properties of Dropout

2015-03-06 · Prateek Jain, Vivek Kulkarni, Abhradeep Thakurta, Oliver Williams

Training deep belief networks (DBNs) requires optimizing a non-convex function with an extremely large number of parameters. Naturally, existing gradient descent (GD) based methods are prone to arbitrarily poor local min…

L2 Regularization

Explaining Landscape Connectivity of Low-cost Solutions for Multilayer Nets

2019-06-14 · NeurIPS 2019 12 · Rohith Kuditipudi, Xiang Wang, Holden Lee, Yi Zhang 외

Mode connectivity is a surprising phenomenon in the loss landscape of deep nets. Optima -- at least those discovered by gradient-based optimization -- turn out to be connected by simple paths on which the loss function i…

The Implicit and Explicit Regularization Effects of Dropout

2020-02-28 · ICML 2020 1 · Colin Wei, Sham Kakade, Tengyu Ma

Dropout is a widely-used regularization technique, often required to obtain state-of-the-art for a number of architectures. This work demonstrates that dropout introduces two distinct but entangled regularization effects…

On the Inductive Bias of Dropout

2014-12-15 · David P. Helmbold, Philip M. Long

Dropout is a simple but effective technique for learning in neural networks and other settings. A sound theoretical understanding of dropout is needed to determine when dropout should be applied and how to use it most ef…

Inductive Bias

Understanding Dropout

2013-12-01 · NeurIPS 2013 12 · Pierre Baldi, Peter J. Sadowski

Dropout is a relatively new algorithm for training neural networks which relies on stochastically dropping out'' neurons during training in order to avoid the co-adaptation of feature detectors. We introduce a general fo…