Diversifying Deep Ensembles: A Saliency Map Approach for Enhanced OOD Detection, Calibration, and Accuracy

TL;DR

Through incorporating saliency map diversification, the proposed Saliency Diversified Deep Ensemble (SDDE) method outperforms conventional ensemble techniques and improves calibration in multiple classification and OOD detection tasks.

Abstract

Deep ensembles are capable of achieving state-of-the-art results in classification and out-of-distribution (OOD) detection. However, their effectiveness is limited due to the homogeneity of learned patterns within ensembles. To overcome this issue, our study introduces Saliency Diversified Deep Ensemble (SDDE), a novel approach that promotes diversity among ensemble members by leveraging saliency maps. Through incorporating saliency map diversification, our method outperforms conventional ensemble techniques and improves calibration in multiple classification and OOD detection tasks. In particular, the proposed method achieves state-of-the-art OOD detection quality, calibration, and accuracy on multiple benchmarks, including CIFAR10/100 and large-scale ImageNet datasets.

Figures (6)

• Figure 1: Saliency diversification. Compared to Deep Ensembles, the models within the proposed SDDE ensemble use different features for prediction, leading to improved generalization and confidence estimation.
• Figure 2: The dependency of ensemble predictions agreement on saliency maps cosine similarity. Saliency maps are computed using GradCAM. Mean and STD values w.r.t. multiple training seeds are reported.
• Figure 3: The training pipeline, where we compute saliency maps using the GradCAM method for each model and apply a diversity loss. The final loss also includes the cross-entropy loss.
• Figure 4: Class Activation Maps (CAMs) for SDDE and the baseline methods. SDDE increases the diversity of CAMs by focusing on different regions of the images.
• Figure 5: Pairwise distributions of cosine similarities between Class Activation Maps (CAMs) of ensemble models.