#regularization
14 resultsComponent-based regularisation of multivariate generalised linear mixed models
Jocelyn Chauvet, Catherine Trottier, Xavier Bry
The paper proposes a component-based regularization method for multivariate generalized linear mixed models with many redundant explanatory variables, using orthogonal components t…
Renormalisation in Quantum Field Theory
Sunil Mukhi
This paper offers an introductory review of renormalisation methods in quantum field theory, compiled from lecture notes delivered at a school in 2017.
Top performing stocks recommendation strategy for portfolio
Kartikay Gupta, Niladri Chatterjee
The paper proposes a regression-based approach that outputs scores for ranking stocks, aiming to recommend top-performing stocks on Indian exchanges, and evaluates the method on tw…
Group Pruning using a Bounded-Lp norm for Group Gating and Regularization
Chaithanya Kumar Mummadi, Tim Genewein, Dan Zhang +2
The paper proposes a gating mechanism and a bounded L1 regularizer to enable group-wise channel pruning in convolutional neural networks, achieving significant parameter reductions…
Adaptive Weight Decay for Deep Neural Networks
Kensuke Nakamura, Byung-Woo Hong
The paper proposes AdaDecay, an adaptive weight-decay technique that adjusts regularization per parameter based on gradient magnitude, and demonstrates improved generalization on M…
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
Sangdoo Yun, Dongyoon Han, Seong Joon Oh +3
The paper introduces CutMix, a data augmentation method that cuts patches from one image and pastes them onto another while mixing their labels, improving classification, localizat…
A L2-norm regularized incremental-stencil WENO scheme for compressible flows
Yujie Zhu, Xiangyu Hu
The paper introduces an L2-norm regularized incremental-stencil WENO scheme to improve robustness and reduce numerical dissipation in compressible flow simulations, using adaptive…
Metric Learning With HORDE: High-Order Regularizer for Deep Embeddings
Pierre Jacob, David Picard, Aymeric Histace +1
The paper introduces HORDE, a distribution-aware regularizer that reduces scattering of deep image features by enforcing locally consistent feature distributions, improving metric…
Self-Balanced Dropout
Shen Li, Chenhao Su, Renfen Hu +1
The paper demonstrates that conventional dropout does not fully eliminate co-adaptation due to input correlations and introduces Self-Balanced Dropout, a trainable dropout variant…
Sobolev Descent
Youssef Mroueh, Tom Sercu, Anant Raj
The paper introduces Sobolev Descent, a method that simplifies GAN training by transporting particles from a source to a target distribution using gradient flows of a Sobolev GAN c…
Electroweak Current Operators in Chiral Effective Field Theory
Hermann Krebs
The paper reviews the construction of nuclear electroweak current operators in chiral effective field theory, highlighting how gauge and chiral symmetries shape continuity equation…
Differential Privacy for Sparse Classification Learning
Puyu Wang, Hai Zhang
The paper proposes a differentially private framework for sparse classification using ADMM, adding exponential noise to stable algorithm steps and providing theoretical privacy bou…
Space-adaptive anisotropic bivariate Laplacian regularization for image restoration
Luca Calatroni, Alessandro Lanza, Monica Pragliola +1
The paper proposes a space‑variant anisotropic bivariate Laplacian regularizer for variational image restoration, estimating its parameters via maximum likelihood and solving the r…
Beyond the standard model with sum rules
Damian Ejlli
The paper derives sum‑rule constraints on particle degrees of freedom and masses from requiring a finite, Lorentz‑invariant zero‑point stress‑energy tensor, and examines how these…