Uncertainty Injection

Uncertainty injection is a technique used to enhance the robustness and generalization of machine learning models, particularly deep learning models, by explicitly incorporating uncertainty into the training process. Current research focuses on applying this technique to various optimization problems, including resource allocation in communication networks and semi-supervised semantic segmentation in image analysis, often employing deep neural networks and Bayesian optimization methods. This approach addresses the limitations of traditional methods that assume perfect data, leading to improved model performance and reliability in real-world applications characterized by noisy or incomplete information. The resulting robust models are valuable across diverse fields, from optimizing wireless communication systems to improving medical image analysis.

Papers

December 3, 2024

GUESS: Generative Uncertainty Ensemble for Self Supervision
Salman Mohamadi, Gianfranco Doretto, Donald A. Adjeroh
Self Supervised Self Supervision Robust Representation Uncertainty Representation Greedy Sampling Uncertainty Injection

June 5, 2024

Robust Communication and Computation using Deep Learning via Joint Uncertainty Injection
Robert-Jeron Reifert, Hayssam Dahrouj, Alaa Alameer Ahmad, Haris Gacanin, Aydin Sezgin
Deep Learning Deep Neural Network Computation Method Robust Communication Delay Resilient Computational Uncertainty Uncertainty Injection

November 30, 2023

Semi-supervised Semantic Segmentation via Boosting Uncertainty on Unlabeled Data
Daoan Zhang, Yunhao Luo, Jianguo Zhang
High Uncertainty Anticipation Unlabeled Data Uncertainty Aware Semi Supervised Semantic Segmentation Uncertainty Injection

February 23, 2023

Uncertainty Injection: A Deep Learning Method for Robust Optimization
Wei Cui, Wei Yu
Deep Learning Robust Optimization Power Control Robust Optimization Problem Robust Solution Uncertainty Injection

July 22, 2022

On Controller Tuning with Time-Varying Bayesian Optimization
Paul Brunzema, Alexander von Rohr, Sebastian Trimpe
Gaussian Process Optimal Control Controller Tuning Uncertainty Injection

November 23, 2021

Uncertainty-Aware Deep Co-training for Semi-supervised Medical Image Segmentation
Xu Zheng, Chong Fu, Haoyu Xie, Jialei Chen, Xingwei Wang, Chiu-Wing Sham
Semi Supervised Learning Semi Supervised Semi Supervised Medical Image Segmentation Co Training Uncertainty Map Pixel Wise Annotated Uncertainty Injection