Why resnet works? residuals generalize
Residual connections significantly boost the performance of deep neural networks.
However, few theoretical results address the influence of residuals on the hypothesis …
However, few theoretical results address the influence of residuals on the hypothesis …
Stable and fair classification
L Huang, N Vishnoi - International Conference on Machine …, 2019 - proceedings.mlr.press
In a recent study, Friedler et al. pointed out that several fair classification algorithms are not
stable with respect to variations in the training set–a crucial consideration in several …
stable with respect to variations in the training set–a crucial consideration in several …
[HTML][HTML] Overfit detection method for deep neural networks trained to beamform ultrasound images
Deep neural networks (DNNs) have remarkable potential to reconstruct ultrasound images.
However, this promise can suffer from overfitting to training data, which is typically detected …
However, this promise can suffer from overfitting to training data, which is typically detected …
Stability and generalization for randomized coordinate descent
Randomized coordinate descent (RCD) is a popular optimization algorithm with wide
applications in solving various machine learning problems, which motivates a lot of …
applications in solving various machine learning problems, which motivates a lot of …
Characterization of excess risk for locally strongly convex population risk
We establish upper bounds for the expected excess risk of models trained by proper iterative
algorithms which approximate the local minima. Unlike the results built upon the strong …
algorithms which approximate the local minima. Unlike the results built upon the strong …
Subsampling oriented active learning method for multi-category classification problem.
SHI Wei, H Honglan, F Yanghe… - Systems Engineering …, 2021 - search.ebscohost.com
Because the computational amount of the traditional active learning method increases
exponentially with the increase of problem size, it is difficult to apply to the large-scale multi …
exponentially with the increase of problem size, it is difficult to apply to the large-scale multi …
The Impact of the Mini-batch Size on the Dynamics of SGD: Variance and Beyond
We study mini-batch stochastic gradient descent (SGD) dynamics under linear regression
and deep linear networks by focusing on the variance of the gradients only given the initial …
and deep linear networks by focusing on the variance of the gradients only given the initial …
[ספר][B] Efficient Stochastic Optimization Algorithms for Large-Scale Machine Learning Problems
C Tan - 2019 - search.proquest.com
Due to rapid growth in the data size, it becomes a more and more challenging issue
concerning how to train machine learning models efficiently. In many industry applications, it …
concerning how to train machine learning models efficiently. In many industry applications, it …
[PDF][PDF] Exploring the Generalization Performance of Neural Networks via Diversity.
Neural networks (NNs) have achieved excellent performance in many industrial tasks, but
their interpretability is still a major challenge and difficulty, in which the generalization ability …
their interpretability is still a major challenge and difficulty, in which the generalization ability …
[CITATION][C] Challenges of statistical machine learning when encountered deep neural network
Q Meng - 2024 - Elsevier