Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:1168-1176, 2016.
Abstract
While the authors of Batch Normalization (BN) identify and address an important problem involved in training deep networks– \textitInternal Covariate Shift– the current solution has certain drawbacks. For instance, BN depends on batch statistics for layerwise input normalization during training which makes the estimates of mean and standard deviation of input (distribution) to hidden layers inaccurate due to shifting parameter values (especially during initial training epochs). Another fundamental problem with BN is that it cannot be used with batch-size 1 during training. We address these drawbacks of BN by proposing a non-adaptive normalization technique for removing covariate shift, that we call \textitNormalization Propagation. Our approach does not depend on batch statistics, but rather uses a data-independent parametric estimate of mean and standard-deviation in every layer thus being computationally faster compared with BN. We exploit the observation that the pre-activation before Rectified Linear Units follow Gaussian distribution in deep networks, and that once the first and second order statistics of any given dataset are normalized, we can forward propagate this normalization without the need for recalculating the approximate statistics for hidden layers.
@InProceedings{pmlr-v48-arpitb16,
title = {Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks},
author = {Devansh Arpit and Yingbo Zhou and Bhargava Kota and Venu Govindaraju},
booktitle = {Proceedings of The 33rd International Conference on Machine Learning},
pages = {1168--1176},
year = {2016},
editor = {Maria Florina Balcan and Kilian Q. Weinberger},
volume = {48},
series = {Proceedings of Machine Learning Research},
address = {New York, New York, USA},
month = {20--22 Jun},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v48/arpitb16.pdf},
url = {http://proceedings.mlr.press/v48/arpitb16.html},
abstract = {While the authors of Batch Normalization (BN) identify and address an important problem involved in training deep networks– \textitInternal Covariate Shift– the current solution has certain drawbacks. For instance, BN depends on batch statistics for layerwise input normalization during training which makes the estimates of mean and standard deviation of input (distribution) to hidden layers inaccurate due to shifting parameter values (especially during initial training epochs). Another fundamental problem with BN is that it cannot be used with batch-size 1 during training. We address these drawbacks of BN by proposing a non-adaptive normalization technique for removing covariate shift, that we call \textitNormalization Propagation. Our approach does not depend on batch statistics, but rather uses a data-independent parametric estimate of mean and standard-deviation in every layer thus being computationally faster compared with BN. We exploit the observation that the pre-activation before Rectified Linear Units follow Gaussian distribution in deep networks, and that once the first and second order statistics of any given dataset are normalized, we can forward propagate this normalization without the need for recalculating the approximate statistics for hidden layers.}
}
%0 Conference Paper
%T Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks
%A Devansh Arpit
%A Yingbo Zhou
%A Bhargava Kota
%A Venu Govindaraju
%B Proceedings of The 33rd International Conference on Machine Learning
%C Proceedings of Machine Learning Research
%D 2016
%E Maria Florina Balcan
%E Kilian Q. Weinberger
%F pmlr-v48-arpitb16
%I PMLR
%J Proceedings of Machine Learning Research
%P 1168--1176
%U http://proceedings.mlr.press
%V 48
%W PMLR
%X While the authors of Batch Normalization (BN) identify and address an important problem involved in training deep networks– \textitInternal Covariate Shift– the current solution has certain drawbacks. For instance, BN depends on batch statistics for layerwise input normalization during training which makes the estimates of mean and standard deviation of input (distribution) to hidden layers inaccurate due to shifting parameter values (especially during initial training epochs). Another fundamental problem with BN is that it cannot be used with batch-size 1 during training. We address these drawbacks of BN by proposing a non-adaptive normalization technique for removing covariate shift, that we call \textitNormalization Propagation. Our approach does not depend on batch statistics, but rather uses a data-independent parametric estimate of mean and standard-deviation in every layer thus being computationally faster compared with BN. We exploit the observation that the pre-activation before Rectified Linear Units follow Gaussian distribution in deep networks, and that once the first and second order statistics of any given dataset are normalized, we can forward propagate this normalization without the need for recalculating the approximate statistics for hidden layers.
TY - CPAPER
TI - Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks
AU - Devansh Arpit
AU - Yingbo Zhou
AU - Bhargava Kota
AU - Venu Govindaraju
BT - Proceedings of The 33rd International Conference on Machine Learning
PY - 2016/06/11
DA - 2016/06/11
ED - Maria Florina Balcan
ED - Kilian Q. Weinberger
ID - pmlr-v48-arpitb16
PB - PMLR
SP - 1168
DP - PMLR
EP - 1176
L1 - http://proceedings.mlr.press/v48/arpitb16.pdf
UR - http://proceedings.mlr.press/v48/arpitb16.html
AB - While the authors of Batch Normalization (BN) identify and address an important problem involved in training deep networks– \textitInternal Covariate Shift– the current solution has certain drawbacks. For instance, BN depends on batch statistics for layerwise input normalization during training which makes the estimates of mean and standard deviation of input (distribution) to hidden layers inaccurate due to shifting parameter values (especially during initial training epochs). Another fundamental problem with BN is that it cannot be used with batch-size 1 during training. We address these drawbacks of BN by proposing a non-adaptive normalization technique for removing covariate shift, that we call \textitNormalization Propagation. Our approach does not depend on batch statistics, but rather uses a data-independent parametric estimate of mean and standard-deviation in every layer thus being computationally faster compared with BN. We exploit the observation that the pre-activation before Rectified Linear Units follow Gaussian distribution in deep networks, and that once the first and second order statistics of any given dataset are normalized, we can forward propagate this normalization without the need for recalculating the approximate statistics for hidden layers.
ER -
Arpit, D., Zhou, Y., Kota, B. & Govindaraju, V.. (2016). Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks. Proceedings of The 33rd International Conference on Machine Learning, in PMLR 48:1168-1176
This site last compiled Sat, 04 Nov 2017 20:59:32 +0000