Abstract
AbstractThis paper presents a novel holistic deep learning framework that simultaneously addresses the challenges of vulnerability to input perturbations, overparametrization, and performance instability from different train-validation splits. The proposed framework holistically improves accuracy, robustness, sparsity, and stability over standard deep learning models, as demonstrated by extensive experiments on both tabular and image data sets. The results are further validated by ablation experiments and SHAP value analysis, which reveal the interactions and trade-offs between the different evaluation metrics. To support practitioners applying our framework, we provide a prescriptive approach that offers recommendations for selecting an appropriate training loss function based on their specific objectives. All the code to reproduce the results can be found at https://github.com/kimvc7/HDL.
Funder
Massachusetts Institute of Technology
Publisher
Springer Science and Business Media LLC
Subject
Artificial Intelligence,Software
Reference60 articles.
1. Abadi, M., Agarwal, A., et al. (2015). TensorFlow: Large-scale machine learning on heterogeneous systems. https://www.tensorflow.org/, software available from tensorflow.org.
2. Aghasi, A., Abdi, A., & Romberg, J. (2020). Fast convex pruning of deep neural networks. SIAM Journal on Mathematics of Data Science, 2(1), 158–188.
3. Amram, M., Dunn, J., & Zhuo, Y. D. (2022). Optimal policy trees. Machine Learning, 111, 2741–2768.
4. Anderson, R., Huchette, J., Ma, W., et al. (2020). Strong mixed-integer programming formulations for trained neural networks. Mathematical Programming (pp. 1–37).
5. Athalye, A., Carlini, N., & Wagner, D. (2018). Obfuscated gradients give a false sense of security: Circumventing defenses to adversarial examples. arXiv:1802.00420.
Cited by
3 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献