Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
149 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

SMOOT: Saliency Guided Mask Optimized Online Training (2310.00772v2)

Published 1 Oct 2023 in cs.CV

Abstract: Deep Neural Networks are powerful tools for understanding complex patterns and making decisions. However, their black-box nature impedes a complete understanding of their inner workings. Saliency-Guided Training (SGT) methods try to highlight the prominent features in the model's training based on the output to alleviate this problem. These methods use back-propagation and modified gradients to guide the model toward the most relevant features while keeping the impact on the prediction accuracy negligible. SGT makes the model's final result more interpretable by masking input partially. In this way, considering the model's output, we can infer how each segment of the input affects the output. In the particular case of image as the input, masking is applied to the input pixels. However, the masking strategy and number of pixels which we mask, are considered as a hyperparameter. Appropriate setting of masking strategy can directly affect the model's training. In this paper, we focus on this issue and present our contribution. We propose a novel method to determine the optimal number of masked images based on input, accuracy, and model loss during the training. The strategy prevents information loss which leads to better accuracy values. Also, by integrating the model's performance in the strategy formula, we show that our model represents the salient features more meaningful. Our experimental results demonstrate a substantial improvement in both model accuracy and the prominence of saliency, thereby affirming the effectiveness of our proposed solution.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (47)
  1. Intelligible models for healthcare: Predicting pneumonia risk and hospital 30-day readmission. In Proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining, pages 1721–1730, 2015.
  2. Tell me where to look: Guided attention inference network. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 9215–9223, 2018.
  3. Modeling annotators: A generative approach to learning from annotator rationales. In Proceedings of the 2008 conference on Empirical methods in natural language processing, pages 31–40, 2008.
  4. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PloS one, 10(7):e0130140, 2015.
  5. Investigating the influence of noise and distractors on the interpretation of neural networks. arXiv preprint arXiv:1611.07270, 2016.
  6. Smoothgrad: removing noise by adding noise. arXiv preprint arXiv:1706.03825, 2017.
  7. Understanding impacts of high-order loss approximations and features in deep learning interpretation. In International Conference on Machine Learning, pages 5848–5856. PMLR, 2019.
  8. Hide-and-seek: Forcing a network to be meticulous for weakly-supervised object and action localization. In 2017 IEEE international conference on computer vision (ICCV), pages 3544–3553. IEEE, 2017.
  9. Deep learning. nature, 521(7553):436–444, 2015.
  10. How to explain individual classification decisions. The Journal of Machine Learning Research, 11:1803–1831, 2010.
  11. Visualizing and understanding convolutional networks. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part I 13, pages 818–833. Springer, 2014.
  12. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision, pages 618–626, 2017.
  13. Learning important features through propagating activation differences. In International conference on machine learning, pages 3145–3153. PMLR, 2017.
  14. Towards better understanding of gradient-based attribution methods for deep neural networks. arXiv preprint arXiv:1711.06104, 2017.
  15. Sanity checks for saliency maps. Advances in neural information processing systems, 31, 2018.
  16. Interpretation of neural networks is fragile. In Proceedings of the AAAI conference on artificial intelligence, volume 33, pages 3681–3688, 2019.
  17. Fever: a large-scale dataset for fact extraction and verification. arXiv preprint arXiv:1803.05355, 2018.
  18. Benchmarking deep learning interpretability in time series predictions. Advances in neural information processing systems, 33:6441–6452, 2020.
  19. Axiomatic attribution for deep networks. In International conference on machine learning, pages 3319–3328. PMLR, 2017.
  20. A benchmark for interpretability methods in deep neural networks. Advances in neural information processing systems, 32, 2019.
  21. Right for the right reasons: Training differentiable models by constraining their explanations. arXiv preprint arXiv:1703.03717, 2017.
  22. Distilling a neural network into a soft decision tree. arXiv preprint arXiv:1711.09784, 2017.
  23. Interpretability of deep learning models: A survey of results. In 2017 IEEE smartworld, ubiquitous intelligence & computing, advanced & trusted computed, scalable computing & communications, cloud & big data computing, Internet of people and smart city innovation (smartworld/SCALCOM/UIC/ATC/CBDcom/IOP/SCI), pages 1–6. IEEE, 2017.
  24. Grafting: Fast, incremental feature selection by gradient descent in function space. The Journal of Machine Learning Research, 3:1333–1356, 2003.
  25. Eraser: A benchmark to evaluate rationalized nlp models. arXiv preprint arXiv:1911.03429, 2019.
  26. Saliency learning: Teaching the model where to pay attention. arXiv preprint arXiv:1902.08649, 2019.
  27. Sharpen focus: Learning with attention separability and consistency. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 512–521, 2019.
  28. Improved regularization of convolutional neural networks with cutout. arXiv preprint arXiv:1708.04552, 2017.
  29. Improving deep learning interpretability by saliency guided training. Advances in Neural Information Processing Systems, 34:26726–26739, 2021.
  30. Guided integrated gradients: An adaptive path method for removing noise. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5050–5058, 2021.
  31. Improved and interpretable defense to transferred adversarial examples by jacobian norm with selective input gradient regularization. arXiv preprint arXiv:2207.13036, 2022.
  32. Mnist handwritten digit database. att labs, 2010.
  33. Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747, 2017.
  34. Learning multiple layers of features from tiny images. 2009.
  35. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.
  36. Matthew D Zeiler. Adadelta: an adaptive learning rate method. arXiv preprint arXiv:1212.5701, 2012.
  37. Captum: A unified and generic model interpretability library for pytorch. arXiv preprint arXiv:2009.07896, 2020.
  38. Rise: Randomized input sampling for explanation of black-box models. arXiv preprint arXiv:1806.07421, 2018.
  39. Attention is all you need. Advances in neural information processing systems, 30, 2017.
  40. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805, 2018.
  41. Sanity checks for saliency metrics. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 6021–6029, 2020.
  42. Gpt-4: Technical report. Technical report, OpenAI, San Francisco, CA, USA, July 2023.
  43. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020.
  44. Going deeper with image transformers. In Proceedings of the IEEE/CVF international conference on computer vision, pages 32–42, 2021.
  45. End-to-end object detection with transformers. In European conference on computer vision, pages 213–229. Springer, 2020.
  46. Pre-trained image processing transformer. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 12299–12310, 2021.
  47. Deit iii: Revenge of the vit. In European Conference on Computer Vision, pages 516–533. Springer, 2022.
Citations (9)
View on Semantic Scholar

Summary

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

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