Uncertainty Representation

Uncertainty representation in machine learning aims to quantify and model the inherent uncertainty in predictions, improving model reliability and robustness. Current research focuses on developing methods to represent uncertainty within various model architectures, including Bayesian neural networks, variational autoencoders, and specialized layers like Kalman filters for sequential data, often employing techniques like heteroscedastic regression and stochastic weight averaging. This work is crucial for building trustworthy AI systems across diverse applications, from robotics and autonomous navigation to medical diagnosis and financial modeling, where understanding and managing uncertainty is paramount for safe and effective deployment.

Papers

September 25, 2024

Uncertainty Representations in State-Space Layers for Deep Reinforcement Learning under Partial Observability
Carlos E. Luis, Alessandro G. Bottero, Julia Vinogradska, Felix Berkenkamp, Jan Peters
Deep Reinforcement Learning Kalman Filter Partial Observability Uncertainty Representation Reinforcement Learning Architecture State Space Layer

June 13, 2024

Flexible Heteroscedastic Count Regression with Deep Double Poisson Networks
Spencer Young, Porter Jenkins, Lonchao Da, Jeff Dotson, Hua Wei
Uncertainty Quantification Regression Task Poisson Surface Reconstruction Heteroscedastic Regression Uncertainty Representation

April 23, 2024

April 18, 2024

Quantifying Aleatoric and Epistemic Uncertainty with Proper Scoring Rules
Paul Hofman, Yusuf Sale, Eyke Hüllermeier
Loss Function Epistemic Uncertainty Aleatoric Uncertainty Relative Probability Scoring Rule Non Gaussian Aleatoric Uncertainty Uncertainty Representation

October 16, 2023

Orthogonal Uncertainty Representation of Data Manifold for Robust Long-Tailed Learning
Yanbiao Ma, Licheng Jiao, Fang Liu, Shuyuan Yang, Xu Liu, Lingling Li
Long Tailed Long Tailed Learning Long Tail Long Tailed Distribution Data Manifold Uncertainty Representation Balanced Classification

September 27, 2023

Bayesian Personalized Federated Learning with Shared and Personalized Uncertainty Representations
Hui Chen, Hengyu Liu, Longbing Cao, Tiancheng Zhang
Personalized Federated Learning Bayesian Perspective Hidden Representation Sharing Matter Uncertainty Representation Personalized Prior

June 16, 2023

Is the Volume of a Credal Set a Good Measure for Epistemic Uncertainty?
Yusuf Sale, Michele Caprio, Eyke Hüllermeier
Epistemic Uncertainty Convex Set Automatic Volume Uncertainty Measure Uncertainty Representation Credal Set

April 10, 2023

Uncertainty-Aware Natural Language Inference with Stochastic Weight Averaging
Aarne Talman, Hande Celikkanat, Sami Virpioja, Markus Heinonen, Jörg Tiedemann
Language Understanding Natural Language Inference Bayesian Uncertainty Stochastic Weight Averaging Uncertainty Representation

December 14, 2022

Modeling Multimodal Aleatoric Uncertainty in Segmentation with Mixture of Stochastic Experts
Zhitong Gao, Yucong Chen, Chuyu Zhang, Xuming He
Segmentation Based Approach Mixture Component Aleatoric Uncertainty Segmentation Quality Calibrated Uncertainty Uncertainty Representation Multimodal Uncertainty

April 2, 2022

UrbanFly: Uncertainty-Aware Planning for Navigation Amongst High-Rises with Monocular Visual-Inertial SLAM Maps
Sudarshan S Harithas, Ayyappa Swamy Thatavarthy, Gurkirat Singh, Arun K Singh, K Madhava Krishna
Trajectory Optimization Uncertainty Aware Visual Inertial SLAM Quadrotor Navigation Uncertainty Representation Navigation Scenario Cross City

March 22, 2022

Representation Uncertainty in Self-Supervised Learning as Variational Inference
Hiroki Nakamura, Masashi Okada, Tadahiro Taniguchi
Self Supervised Learning Variational Inference Variational Autoencoder Unsupervised Representation Learning Probabilistic Generative Model Uncertainty Representation

Uncertainty Representation

Papers

Uncertainty Representations in State-Space Layers for Deep Reinforcement Learning under Partial Observability

Flexible Heteroscedastic Count Regression with Deep Double Poisson Networks

Uncertainty in latent representations of variational autoencoders optimized for visual tasks

Uncertainty Quantification on Graph Learning: A Survey

Quantifying Aleatoric and Epistemic Uncertainty with Proper Scoring Rules

Orthogonal Uncertainty Representation of Data Manifold for Robust Long-Tailed Learning

Bayesian Personalized Federated Learning with Shared and Personalized Uncertainty Representations

Is the Volume of a Credal Set a Good Measure for Epistemic Uncertainty?

Uncertainty-Aware Natural Language Inference with Stochastic Weight Averaging

Modeling Multimodal Aleatoric Uncertainty in Segmentation with Mixture of Stochastic Experts

UrbanFly: Uncertainty-Aware Planning for Navigation Amongst High-Rises with Monocular Visual-Inertial SLAM Maps

Representation Uncertainty in Self-Supervised Learning as Variational Inference