Add to ORKGUncertainty estimation (UE) techniques -- such as the Gaussian process (GP), Bayesian neural networks (BNN), Monte Carlo dropout (MCDropout) -- aim to improve the interpretability of machine learning models by assigning an estimated uncertainty value to each of their prediction outputs. However, since too high uncertainty estimates can have fatal consequences in practice, this paper analyzes the above techniques. Firstly, we show that GP methods always yield high uncertainty estimates on out of distribution (OOD) data. Secondly, we show on a 2D toy example that both BNNs and MCDropout do not give high uncertainty estimates on OOD samples. Finally, we show empirically that this pitfall of BNNs and MCDropout holds on real world datasets as w...
As neural networks become more popular, the need for accompanying uncertainty estimates increases. T...
Machine learning model deployment in clinical practice demands real-time risk assessment to identify...
Estimating how uncertain an AI system is in its predictions is important to improve the safety of su...
Deep learning tools have gained tremendous attention in applied machine learning. However such tools...
Neural networks are an important and powerful family of models, but they have lacked practical ways ...
Deep learning (DL), which involves powerful black box predictors, has achieved a remarkable performa...
Deep Learning (DL) has achieved the state-of-the-art performance across a broad spectrum oftasks. Fr...
Accurate and efficient uncertainty estimation is crucial to build reliable Machine Learning (ML) mod...
Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, Engineering Syste...
One major impediment to the wider use of deep learning for clinical decision making is the difficult...
The breakout success of deep neural networks (NNs) in the 2010's marked a new era in the quest to bu...
Both uncertainty estimation and interpretability are important factors for trustworthy machine learn...
Uncertainty plays a crucial role in the machine learning field. Both model trustworthiness and perfo...
Machine learning and artificial intelligence will be deeply embedded in the intelligent systems huma...
Uncertainty quantification for complex deep learning models is increasingly important as these techn...
As neural networks become more popular, the need for accompanying uncertainty estimates increases. T...
Machine learning model deployment in clinical practice demands real-time risk assessment to identify...
Estimating how uncertain an AI system is in its predictions is important to improve the safety of su...
Deep learning tools have gained tremendous attention in applied machine learning. However such tools...
Neural networks are an important and powerful family of models, but they have lacked practical ways ...
Deep learning (DL), which involves powerful black box predictors, has achieved a remarkable performa...
Deep Learning (DL) has achieved the state-of-the-art performance across a broad spectrum oftasks. Fr...
Accurate and efficient uncertainty estimation is crucial to build reliable Machine Learning (ML) mod...
Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, Engineering Syste...
One major impediment to the wider use of deep learning for clinical decision making is the difficult...
The breakout success of deep neural networks (NNs) in the 2010's marked a new era in the quest to bu...
Both uncertainty estimation and interpretability are important factors for trustworthy machine learn...
Uncertainty plays a crucial role in the machine learning field. Both model trustworthiness and perfo...
Machine learning and artificial intelligence will be deeply embedded in the intelligent systems huma...
Uncertainty quantification for complex deep learning models is increasingly important as these techn...
As neural networks become more popular, the need for accompanying uncertainty estimates increases. T...
Machine learning model deployment in clinical practice demands real-time risk assessment to identify...
Estimating how uncertain an AI system is in its predictions is important to improve the safety of su...