AutoDEUQ: Automated Deep Ensemble with Uncertainty Quantification

Romain Egele (1, 2) , Romit Maulik (1) , Krishnan Raghavan (1) , Bethany Lusch (1) , Isabelle Guyon (3, 2) , Prasanna Balaprakash (1)

1 ANL - Argonne National Laboratory [Lemont]
2 LISN - Laboratoire Interdisciplinaire des Sciences du Numérique
3 TAU - TAckling the Underspecified

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.

Domaines: Intelligence artificielle [cs.AI]

Romain Egele, Romit Maulik, Krishnan Raghavan, Bethany Lusch, Isabelle Guyon, et al.. AutoDEUQ: Automated Deep Ensemble with Uncertainty Quantification. 26TH International Conference on Pattern Recognition, Aug 2022, Montréal, Canada. pp.1908-1914. ⟨hal-03518597⟩