Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Diversifying Deep Ensembles: A Saliency Map Approach for Enhanced OOD Detection, Calibration, and Accuracy (2305.11616v5)

Published 19 May 2023 in cs.CV, cs.AI, and cs.LG

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.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (37)
  1. Deep ensembles work, but are they necessary? Advances in Neural Information Processing Systems, 35:33646–33660, 2022.
  2. Matti Aksela. Comparison of classifier selection methods for improving committee performance. In International Workshop on Multiple Classifier Systems, pp. 84–93. Springer, 2003.
  3. Pitfalls of in-domain uncertainty estimation and ensembling in deep learning. arXiv preprint arXiv:2002.06470, 2020.
  4. Are we done with imagenet? arXiv preprint arXiv:2006.07159, 2020.
  5. Glenn W Brier et al. Verification of forecasts expressed in terms of probability. Monthly weather review, 78(1):1–3, 1950.
  6. Arcface: Additive angular margin loss for deep face recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  4690–4699, 2019.
  7. Bayesian neural networks: An introduction and survey. Case Studies in Applied Bayesian Data Science: CIRM Jean-Morlet Chair, Fall 2018, pp.  45–87, 2020.
  8. Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572, 2014.
  9. In search of lost domain generalization. ArXiv, abs/2007.01434, 2020.
  10. On calibration of modern neural networks. In International conference on machine learning, pp. 1321–1330. PMLR, 2017.
  11. Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In Proceedings of the IEEE international conference on computer vision, pp.  1026–1034, 2015.
  12. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  770–778, 2016.
  13. Benchmarking neural network robustness to common corruptions and perturbations. Proceedings of the International Conference on Learning Representations, 2019.
  14. A baseline for detecting misclassified and out-of-distribution examples in neural networks. arXiv preprint arXiv:1610.02136, 2016.
  15. Deep anomaly detection with outlier exposure. arXiv preprint arXiv:1812.04606, 2018.
  16. Scaling out-of-distribution detection for real-world settings. arXiv preprint arXiv:1911.11132, 2019.
  17. WILDS: A benchmark of in-the-wild distribution shifts. arXiv, 2020.
  18. Learning multiple layers of features from tiny images. Technical Report 0, University of Toronto, Toronto, Ontario, 2009.
  19. Simple and scalable predictive uncertainty estimation using deep ensembles. Advances in neural information processing systems, 30, 2017.
  20. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324, 1998.
  21. Deep learning for generic object detection: A survey. International journal of computer vision, 128:261–318, 2020.
  22. Sgdr: Stochastic gradient descent with warm restarts. arXiv preprint arXiv:1608.03983, 2016.
  23. Shifts: A dataset of real distributional shift across multiple large-scale tasks. In NeurIPS Dataset & Benchmark Track, 2021.
  24. Shifts 2.0: Extending the dataset of real distributional shifts, 2022.
  25. A metric learning reality check. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXV 16, pp.  681–699. Springer, 2020.
  26. Obtaining well calibrated probabilities using bayesian binning. In Proceedings of the AAAI conference on artificial intelligence, volume 29, 2015.
  27. Improving adversarial robustness via promoting ensemble diversity. In International Conference on Machine Learning, pp. 4970–4979. PMLR, 2019.
  28. Dice: Diversity in deep ensembles via conditional redundancy adversarial estimation. In ICLR 2021-9th International Conference on Learning Representations, 2021.
  29. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision, pp.  618–626, 2017.
  30. Diversity regularization in deep ensembles. arXiv preprint arXiv:1802.07881, 2018.
  31. Deep inside convolutional networks: Visualising image classification models and saliency maps. arXiv preprint arXiv:1312.6034, 2013.
  32. Diverse ensembles improve calibration. arXiv preprint arXiv:2007.04206, 2020.
  33. React: Out-of-distribution detection with rectified activations. Advances in Neural Information Processing Systems, 34:144–157, 2021.
  34. Out-of-distribution detection with deep nearest neighbors. In International Conference on Machine Learning, pp. 20827–20840. PMLR, 2022.
  35. Wild-time: A benchmark of in-the-wild distribution shift over time. In Thirty-sixth Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2022.
  36. Openood v1.5: Enhanced benchmark for out-of-distribution detection. arXiv preprint arXiv:2306.09301, 2023.
  37. Learning deep features for discriminative localization. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  2921–2929, 2016.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now