Noise Based

Noise, inherent in data and hardware, significantly impacts the performance and reliability of machine learning models. Current research focuses on developing robust algorithms and architectures that mitigate the effects of noise, including techniques like explainable regularizations for analog neural networks, noise-aware federated learning strategies, and non-convex loss functions for gradient boosting. These advancements aim to improve model accuracy, generalization, and reproducibility across diverse applications, from healthcare and finance to autonomous systems, by addressing various noise sources and their interactions with data characteristics. The ultimate goal is to build more reliable and trustworthy machine learning systems.

Papers

September 13, 2024

Improving Analog Neural Network Robustness: A Noise-Agnostic Approach with Explainable Regularizations
Alice Duque, Pedro Freire, Egor Manuylovich, Dmitrii Stoliarov, Jaroslaw Prilepsky, Sergei Turitsyn
Deep Learning Architecture Noise Robustness Analog Neural Network Noise Tolerant Network Noise Based Explanation Regularization

June 5, 2024

Noise-Aware Algorithm for Heterogeneous Differentially Private Federated Learning
Saber Malekmohammadi, Yaoliang Yu, Yang Cao
Differential Privacy Robust Version Privacy Analysis Privacy Parameter Differentially Private Federated Learning Noise Based

October 8, 2023

Robust-GBDT: GBDT with Nonconvex Loss for Tabular Classification in the Presence of Label Noise and Class Imbalance
Jiaqi Luo, Yuedong Quan, Shixin Xu
Label Noise Class Imbalance Convex Loss Tabular Classification Noise Based

October 5, 2023

Towards Robust and Generalizable Training: An Empirical Study of Noisy Slot Filling for Input Perturbations
Jiachi Liu, Liwen Wang, Guanting Dong, Xiaoshuai Song, Zechen Wang, Zhengyang Wang, Shanglin Lei, Jinzheng Zhao, Keqing He, Bo Xiao, Weiran Xu
Empirical Study Slot Filling Real World Noisy Datasets Learning Generalizable Input Perturbation Noise Based

June 17, 2023

Adaptive Strategies in Non-convex Optimization
Zhenxun Zhuang
Stochastic Gradient Non Convex Optimization Adaptive Algorithm Adaptive Strategy Gradient Scaling Noise Based

June 14, 2023

Causal Discovery from Time Series with Hybrids of Constraint-Based and Noise-Based Algorithms
Daria Bystrova, Charles K. Assaad, Julyan Arbel, Emilie Devijver, Eric Gaussier, Wilfried Thuiller
Time Series Causal Discovery Causal Graph Efficient Hybrid Constraint Programming Hybrid Method Noise Based

February 8, 2023

Towards Inferential Reproducibility of Machine Learning Research
Michael Hagmann, Philipp Meier, Stefan Riezler
Machine Learning Research Reproducibility Noisy Measurement Noise Based

October 2, 2022

iCTGAN--An Attack Mitigation Technique for Random-vector Attack on Accelerometer-based Gait Authentication Systems
Jun Hyung Mo, Rajesh Kumar
Generative Adversarial Network Gait Dataset Noise Based Based Gait

Noise Based

Papers

Improving Analog Neural Network Robustness: A Noise-Agnostic Approach with Explainable Regularizations

Noise-Aware Algorithm for Heterogeneous Differentially Private Federated Learning

Robust-GBDT: GBDT with Nonconvex Loss for Tabular Classification in the Presence of Label Noise and Class Imbalance

Towards Robust and Generalizable Training: An Empirical Study of Noisy Slot Filling for Input Perturbations

Adaptive Strategies in Non-convex Optimization

Causal Discovery from Time Series with Hybrids of Constraint-Based and Noise-Based Algorithms

Towards Inferential Reproducibility of Machine Learning Research

iCTGAN--An Attack Mitigation Technique for Random-vector Attack on Accelerometer-based Gait Authentication Systems