Mathematical Algorithm Design for Deep Learning under Societal and Judicial Constraints: The Algorithmic Transparency Requirement
Not on the map. Why?
Abstract (arXiv)
Deep learning still has drawbacks in terms of trustworthiness, which describes a comprehensible, fair, safe, and reliable method. To mitigate the potential risk of AI, clear obligations associated to trustworthiness have been proposed via regulatory guidelines, e.g., in the European AI Act. Therefore, a central question is to what extent trustworthy deep learning can be realized. Establishing the described properties constituting trustworthiness requires that the factors influencing an algorithmic computation can be retraced, i.e., the algorithmic implementation is transparent. Motivated by the observation that the current evolution of deep learning models necessitates a change in computing technology, we derive a mathematical framework which enables us to analyze whether a transparent implementation in a computing model is feasible. We exemplarily apply our trustworthiness framework to analyze deep learning approaches for inverse problems in digital and analog computing models represented by Turing and Blum-Shub-Smale Machines, respectively. Based on previous results, we find that Blum-Shub-Smale Machines have the potential to establish trustworthy solvers for inverse problems under fairly general conditions, whereas Turing machines cannot guarantee trustworthiness to the same degree.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 55 of its 153 references that are in the map. First posted on arXiv 18 Jan 2024.
Builds on
55 of 153 references.
References · 55
- 2017Attention Is All You Needcited by 59,932
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2020Language Models are Few-Shot Learnerscited by 32,199
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2013Intriguing properties of neural networkscited by 5,114
- 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
- 2021On the Opportunities and Risks of Foundation Modelscited by 3,961
- 2017Deep reinforcement learning from human preferencescited by 3,846
- 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
- 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
- 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
38 more
- 2023A Survey on Evaluation of Large Language Modelscited by 1,227
- 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
- 2018Robustness May Be at Odds with Accuracycited by 873
- 2016The Mythos of Model Interpretabilitycited by 716
- 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 695
- 2016Error bounds for approximations with deep ReLU networkscited by 567
- 2018Explaining Explanations: An Overview of Interpretability of Machine Learningcited by 508
- 2018Learning to See in the Darkcited by 459
- 2017Learning a Variational Network for Reconstruction of Accelerated MRI Datacited by 340
- 2020Deep Learning Techniques for Inverse Problems in Imagingcited by 290
- 2018Audio Adversarial Examples: Targeted Attacks on Speech-to-Textcited by 268
- 2023A Cookbook of Self-Supervised Learningcited by 248
- 2017Image reconstruction by domain transform manifold learningcited by 231
- 2021A Survey of Human-in-the-loop for Machine Learningcited by 214
- 2019On instabilities of deep learning in image reconstruction - Does AI come at a cost?cited by 211
- 2019Towards Stable and Efficient Training of Verifiably Robust Neural Networkscited by 204
- 2023TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning with Hardware Support for Embeddingscited by 198
- 2020The Computational Limits of Deep Learningcited by 188
- 2015The Power of Depth for Feedforward Neural Networkscited by 181
- 2016Why and When Can Deep -- but Not Shallow -- Networks Avoid the Curse of Dimensionality: a Reviewcited by 155
- 2020Explainable Deep Learning: A Field Guide for the Uninitiatedcited by 127
- 2021Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systemscited by 125
- 2017Optimal Approximation with Sparsely Connected Deep Neural Networkscited by 90
- 2020Privacy in Deep Learning: A Surveycited by 70
- 2021The Modern Mathematics of Deep Learningcited by 63
- 2020The gap between theory and practice in function approximation with deep neural networkscited by 53
- 2021Can stable and accurate neural networks be computed? -- On the barriers of deep learning and Smale's 18th problemcited by 53
- 2019Approximation spaces of deep neural networkscited by 51
- 2019In-memory hyperdimensional computingcited by 47
- 2019Nonlinear Approximation and (Deep) ReLU Networkscited by 45
- 2021A Long Short-Term Memory for AI Applications in Spike-based Neuromorphic Hardwarecited by 32
- 2020Safety Concerns and Mitigation Approaches Regarding the Use of Deep Learning in Safety-Critical Perception Taskscited by 31
- 2020Algorithmic Fairness from a Non-ideal Perspectivecited by 30
- 2018Learning The Invisible: A Hybrid Deep Learning-Shearlet Framework for Limited Angle Computed Tomographycited by 29
- 2023A Survey on Deep Learning Hardware Accelerators for Heterogeneous HPC Platformscited by 21
- 2020Robustness Certification of Generative Modelscited by 19
- 2018Sparse Regularization via Convex Analysiscited by 16
- 2021The mathematics of adversarial attacks in AI -- Why deep learning is unstable despite the existence of stable neural networkscited by 16
50 most cited of 55 All 55 (JSON)
Built on by
Citing papers per year: 2024: 1.
1 paper.