Timezone: »
Double descent refers to the phase transition that is exhibited by the generalization error of unregularized learning models when varying the ratio between the number of parameters and the number of training samples. The recent success of highly over-parameterized machine learning models such as deep neural networks has motivated a theoretical analysis of the double descent phenomenon in classical models such as linear regression which can also generalize well in the over-parameterized regime. We provide the first exact non-asymptotic expressions for double descent of the minimum norm linear estimator. Our approach involves constructing a special determinantal point process which we call surrogate random design, to replace the standard i.i.d. design of the training sample. This surrogate design admits exact expressions for the mean squared error of the estimator while preserving the key properties of the standard design. We also establish an exact implicit regularization result for over-parameterized training samples. In particular, we show that, for the surrogate design, the implicit bias of the unregularized minimum norm estimator precisely corresponds to solving a ridge-regularized least squares problem on the population distribution. In our analysis we introduce a new mathematical tool of independent interest: the class of random matrices for which determinant commutes with expectation.
Author Information
Michal Derezinski (UC Berkeley)
Feynman Liang (Berkeley)
Michael W Mahoney (UC Berkeley)
More from the Same Authors
-
2020 Poster: Boundary thickness and robustness in learning models »
Yaoqing Yang · Rajiv Khanna · Yaodong Yu · Amir Gholami · Kurt Keutzer · Joseph Gonzalez · Kannan Ramchandran · Michael W Mahoney -
2020 Poster: Debiasing Distributed Second Order Optimization with Surrogate Sketching and Scaled Regularization »
Michal Derezinski · Burak Bartan · Mert Pilanci · Michael W Mahoney -
2020 Poster: Sampling from a k-DPP without looking at all items »
Daniele Calandriello · Michal Derezinski · Michal Valko -
2020 Spotlight: Sampling from a k-DPP without looking at all items »
Daniele Calandriello · Michal Derezinski · Michal Valko -
2020 Poster: Improved guarantees and a multiple-descent curve for Column Subset Selection and the Nystrom method »
Michal Derezinski · Rajiv Khanna · Michael W Mahoney -
2020 Poster: Precise expressions for random projections: Low-rank approximation and randomized Newton »
Michal Derezinski · Feynman Liang · Zhenyu Liao · Michael W Mahoney -
2020 Oral: Improved guarantees and a multiple-descent curve for Column Subset Selection and the Nystrom method »
Michal Derezinski · Rajiv Khanna · Michael W Mahoney -
2020 Poster: A random matrix analysis of random Fourier features: beyond the Gaussian kernel, a precise phase transition, and the corresponding double descent »
Zhenyu Liao · Romain Couillet · Michael W Mahoney -
2020 Poster: A Statistical Framework for Low-bitwidth Training of Deep Neural Networks »
Jianfei Chen · Yu Gai · Zhewei Yao · Michael W Mahoney · Joseph Gonzalez -
2019 Workshop: Beyond first order methods in machine learning systems »
Anastasios Kyrillidis · Albert Berahas · Fred Roosta · Michael W Mahoney -
2019 Poster: ANODEV2: A Coupled Neural ODE Framework »
Tianjun Zhang · Zhewei Yao · Amir Gholami · Joseph Gonzalez · Kurt Keutzer · Michael W Mahoney · George Biros -
2019 Poster: Distributed estimation of the inverse Hessian by determinantal averaging »
Michal Derezinski · Michael W Mahoney -
2019 Poster: Exact sampling of determinantal point processes with sublinear time preprocessing »
Michal Derezinski · Daniele Calandriello · Michal Valko -
2018 Poster: GIANT: Globally Improved Approximate Newton Method for Distributed Optimization »
Shusen Wang · Fred Roosta · Peng Xu · Michael W Mahoney -
2018 Poster: Leveraged volume sampling for linear regression »
Michal Derezinski · Manfred K. Warmuth · Daniel Hsu -
2018 Spotlight: Leveraged volume sampling for linear regression »
Michal Derezinski · Manfred K. Warmuth · Daniel Hsu -
2018 Poster: Hessian-based Analysis of Large Batch Training and Robustness to Adversaries »
Zhewei Yao · Amir Gholami · Qi Lei · Kurt Keutzer · Michael W Mahoney -
2017 Poster: Unbiased estimates for linear regression via volume sampling »
Michal Derezinski · Manfred K. Warmuth -
2017 Spotlight: Unbiased estimates for linear regression via volume sampling »
Michal Derezinski · Manfred K. Warmuth -
2016 Poster: Feature-distributed sparse regression: a screen-and-clean approach »
Jiyan Yang · Michael W Mahoney · Michael Saunders · Yuekai Sun -
2016 Poster: Sub-sampled Newton Methods with Non-uniform Sampling »
Peng Xu · Jiyan Yang · Farbod Roosta-Khorasani · Christopher RĂ© · Michael W Mahoney -
2016 Poster: Yggdrasil: An Optimized System for Training Deep Decision Trees at Scale »
Firas Abuzaid · Joseph K Bradley · Feynman Liang · Andrew Feng · Lee Yang · Matei Zaharia · Ameet S Talwalkar -
2015 Poster: Fast Randomized Kernel Ridge Regression with Statistical Guarantees »
Ahmed Alaoui · Michael W Mahoney -
2014 Poster: The limits of squared Euclidean distance regularization »
Michal Derezinski · Manfred K. Warmuth -
2014 Spotlight: The limits of squared Euclidean distance regularization »
Michal Derezinski · Manfred K. Warmuth -
2013 Workshop: Large Scale Matrix Analysis and Inference »
Reza Zadeh · Gunnar Carlsson · Michael Mahoney · Manfred K. Warmuth · Wouter M Koolen · Nati Srebro · Satyen Kale · Malik Magdon-Ismail · Ashish Goel · Matei A Zaharia · David Woodruff · Ioannis Koutis · Benjamin Recht