Model Uncertainty

Model uncertainty, encompassing both aleatoric (irreducible) and epistemic (reducible) sources of error, is a critical area of research aiming to improve the reliability and trustworthiness of machine learning models. Current efforts focus on quantifying and decomposing these uncertainties across various model architectures, including deep neural networks, boosted trees, and Bayesian methods, often employing techniques like Monte Carlo dropout, ensembles, and conformal prediction. Understanding and effectively managing model uncertainty is crucial for building robust and safe AI systems, particularly in high-stakes applications like healthcare, finance, and climate modeling, where reliable predictions are paramount. This improved understanding facilitates better decision-making under uncertainty and enhances the overall trustworthiness of AI-driven insights.

Papers

November 15, 2024

Understanding The Effect Of Temperature On Alignment With Human Opinions
Maja Pavlovic, Massimo Poesio
Mixed Effect Alignment Problem Model Uncertainty Human Opinion Temperature Measurement Opinion Distribution Subjective Task

November 5, 2024

Uncertainty Quantification for Clinical Outcome Predictions with (Large) Language Models
Zizhang Chen, Peizhao Li, Xiaomeng Dong, Pengyu Hong
Language Model Uncertainty Quantification Model Uncertainty Clinical Prediction Task Clinical Outcome Prediction

October 14, 2024

Intelligent prospector v2.0: exploration drill planning under epistemic model uncertainty
John Mern, Anthony Corso, Damian Burch, Kurt House, Jef Caers
Model Uncertainty Scientific Hypothesis Geophysical Data Exploration Planner Intelligent Exploration

October 10, 2024

Towards Trustworthy Web Attack Detection: An Uncertainty-Aware Ensemble Deep Kernel Learning Model
Yonghang Zhou, Hongyi Zhu, Yidong Chai, Yuanchun Jiang, Yezheng Liu
Model Uncertainty Uncertainty Aware Deep Learning Web Attack Ensemble Kernel

October 8, 2024

Uncertainty estimation via ensembles of deep learning models and dropout layers for seismic traces
Giovanni Messuti, ortensia Amoroso, Ferdinando Napolitano, Mariarosaria Falanga, Paolo Capuano, Silvia Scarpetta
Deep Learning Convolutional Neural Network Deep Learning Model Uncertainty Estimation Diverse Ensemble Model Uncertainty Dropout Layer

September 20, 2024

Relationship between Uncertainty in DNNs and Adversarial Attacks
Abigail Adeniran, Adewale Adeyemo
Neural Network Deep Neural Network Adversarial Attack High Uncertainty Anticipation DNN Framework Model Uncertainty Human Relationship

September 12, 2024

DEMAU: Decompose, Explore, Model and Analyse Uncertainties
Arthur Hoarau, Vincent Lemaire
Machine Learning Full Model Active Learning High Uncertainty Anticipation Uncertainty Estimation Model Uncertainty Microbial Decomposition High Uncertainty Value

September 5, 2024

Distributionally Robust Optimisation with Bayesian Ambiguity Sets
Charita Dellaporta, Patrick O'Hara, Theodoros Damoulas
Model Uncertainty Distributionally Robust Optimization Ambiguity Detection Ambiguity Set

August 19, 2024

Branch and Bound to Assess Stability of Regression Coefficients in Uncertain Models
Brian Knaeble, R. Mitchell Hughes, George Rudolph, Mark A. Abramson, Daniel Razo
Core Stability Model Uncertainty Regression Model Multi Channel Branch Uncertain Model

August 13, 2024

July 14, 2024

Learning to Steer Markovian Agents under Model Uncertainty
Jiawei Huang, Vinzenz Thoma, Zebang Shen, Heinrich H. Nax, Niao He
LeArning Abstract Multi Agent System Model Uncertainty Learning Dynamic Episodic Reinforcement Learning Steering Technique Markovian Policy

June 26, 2024

Unified Uncertainties: Combining Input, Data and Model Uncertainty into a Single Formulation
Matias Valdenegro-Toro, Ivo Pascal de Jong, Marco Zullich
Raw Data High Uncertainty Anticipation Model Uncertainty Decision Boundary Rich Input Prediction Uncertainty New Formulation Input Uncertainty Unified Uncertainty

June 13, 2024

SeMOPO: Learning High-quality Model and Policy from Low-quality Offline Visual Datasets
Shenghua Wan, Ziyuan Chen, Le Gan, Shuai Feng, De-Chuan Zhan
Model Uncertainty Health Policy Recommendation Offline Data Learned Dynamic Offline Policy

June 11, 2024

Beyond the Norms: Detecting Prediction Errors in Regression Models
Andres Altieri, Marco Romanelli, Georg Pichler, Florence Alberge, Pablo Piantanida
Machine Learning Uncertainty Quantification Probabilistic Model Model Uncertainty Regression Model Aleatoric Uncertainty Prediction Error Learning Norm

May 28, 2024

Determining Domain of Machine Learning Models using Kernel Density Estimates: Applications in Materials Property Prediction
Lane E. Schultz, Yiqi Wang, Ryan Jacobs, Dane Morgan
Machine Learning Model Financial Application Model Uncertainty Kernel Density Material Property Prediction Domain Discriminator Domain Specific Model Material Datasets Significant Dissimilarity

May 27, 2024

Any-step Dynamics Model Improves Future Predictions for Online and Offline Reinforcement Learning
Haoxin Lin, Yu-Yan Xu, Yihao Sun, Zhilong Zhang, Yi-Chen Li, Chengxing Jia, Junyin Ye, Jiaji Zhang, Yang Yu
Reinforcement Learning Offline Reinforcement Learning Model Uncertainty Online Service Sequential Information Future Prediction Step to Step Dynamic

May 25, 2024

LUCIE: A Lightweight Uncoupled ClImate Emulator with long-term stability and physical consistency for O(1000)-member ensembles
Haiwen Guan, Troy Arcomano, Ashesh Chattopadhyay, Romit Maulik
Strong Consistency Diverse Ensemble Model Uncertainty Model Ensemble Based Weather

May 15, 2024

Improving Label Error Detection and Elimination with Uncertainty Quantification
Johannes Jakubik, Michael Vössing, Manil Maskey, Christopher Wölfle, Gerhard Satzger
Uncertainty Quantification Model Uncertainty Label Error Elimination Voting

May 3, 2024

Quantifying Distribution Shifts and Uncertainties for Enhanced Model Robustness in Machine Learning Applications
Vegard Flovik
Synthetic Data Distribution Shift Model Uncertainty Model Robustness Model Generalization Machine Learning Application High Uncertainty Value Model Adaptation

Model Uncertainty

Papers

Understanding The Effect Of Temperature On Alignment With Human Opinions

Uncertainty Quantification for Clinical Outcome Predictions with (Large) Language Models

Intelligent prospector v2.0: exploration drill planning under epistemic model uncertainty

Towards Trustworthy Web Attack Detection: An Uncertainty-Aware Ensemble Deep Kernel Learning Model

Uncertainty estimation via ensembles of deep learning models and dropout layers for seismic traces

Relationship between Uncertainty in DNNs and Adversarial Attacks

DEMAU: Decompose, Explore, Model and Analyse Uncertainties

Distributionally Robust Optimisation with Bayesian Ambiguity Sets

Branch and Bound to Assess Stability of Regression Coefficients in Uncertain Models

Quantification of total uncertainty in the physics-informed reconstruction of CVSim-6 physiology

MAQA: Evaluating Uncertainty Quantification in LLMs Regarding Data Uncertainty

Learning to Steer Markovian Agents under Model Uncertainty

Unified Uncertainties: Combining Input, Data and Model Uncertainty into a Single Formulation

SeMOPO: Learning High-quality Model and Policy from Low-quality Offline Visual Datasets

Beyond the Norms: Detecting Prediction Errors in Regression Models

Determining Domain of Machine Learning Models using Kernel Density Estimates: Applications in Materials Property Prediction

Any-step Dynamics Model Improves Future Predictions for Online and Offline Reinforcement Learning

LUCIE: A Lightweight Uncoupled ClImate Emulator with long-term stability and physical consistency for O(1000)-member ensembles

Improving Label Error Detection and Elimination with Uncertainty Quantification

Quantifying Distribution Shifts and Uncertainties for Enhanced Model Robustness in Machine Learning Applications