AutoDEUQ: Automated Deep Ensemble with Uncertainty Quantification
This work addresses the problem of efficiently generating ensembles for uncertainty estimation in deep learning, which is important for practitioners needing reliable predictions, though it is incremental as it builds on existing ensemble methods.
The authors tackled the challenge of building deep neural network ensembles for uncertainty quantification by proposing AutoDEUQ, an automated approach that uses joint neural architecture and hyperparameter search, and demonstrated it outperforms several existing methods on regression benchmarks.
Deep neural networks are powerful predictors for a variety of tasks. However, they do not capture uncertainty directly. Using neural network ensembles to quantify uncertainty is competitive with approaches based on Bayesian neural networks while benefiting from better computational scalability. However, building ensembles of neural networks is a challenging task because, in addition to choosing the right neural architecture or hyperparameters for each member of the ensemble, there is an added cost of training each model. We propose AutoDEUQ, an automated approach for generating an ensemble of deep neural networks. Our approach leverages joint neural architecture and hyperparameter search to generate ensembles. We use the law of total variance to decompose the predictive variance of deep ensembles into aleatoric (data) and epistemic (model) uncertainties. We show that AutoDEUQ outperforms probabilistic backpropagation, Monte Carlo dropout, deep ensemble, distribution-free ensembles, and hyper ensemble methods on a number of regression benchmarks.