Skip to yearly menu bar Skip to main content

In-Person Oral presentation / top 25% paper

Packed Ensembles for efficient uncertainty estimation

Olivier Laurent · Adrien Lafage · Enzo Tartaglione · Geoffrey Daniel · Jean-marc Martinez · Andrei Bursuc · Gianni Franchi



Deep Ensembles (DE) are a prominent approach for achieving excellent performance on key metrics such as accuracy, calibration, uncertainty estimation, and out-of-distribution detection. However, hardware limitations of real-world systems constrain to smaller ensembles and lower-capacity networks, significantly deteriorating their performance and properties. We introduce Packed-Ensembles (PE), a strategy to design and train lightweight structured ensembles by carefully modulating the dimension of their encoding space. We leverage grouped convolutions to parallelize the ensemble into a single shared backbone and forward pass to improve training and inference speeds. PE is designed to operate within the memory limits of a standard neural network. Our extensive research indicates that PE accurately preserves the properties of DE, such as diversity, and performs equally well in terms of accuracy, calibration, out-of-distribution detection, and robustness to distribution shift. We make our code available at

Chat is not available.