corpus.blog
41554 blogs · [ { "id": "01a08769-d552-72d9-94c7-82667a3efdc7", "title": "Domain-Specific AI Should Focus on Workflows Rather Than Modeling", "url": "https://davidstutz.de/domain-specific-ai-should-focus-on-workflows-rather-than-modeling/", "published_at": "2026-06-14T14:26:01+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667b1b1478", "title": "AI Evaluation is Becoming an Exciting Standalone Discipline", "url": "https://davidstutz.de/ai-evaluation-is-becoming-an-exciting-standalone-discipline/", "published_at": "2026-04-16T18:59:01+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667bf86b5e", "title": "RAISE 2025 Panel Statement on Aligning AI to Clinical Values", "url": "https://davidstutz.de/raise-2025-panel-statement-on-aligning-ai-to-clinical-values/", "published_at": "2025-10-04T21:24:33+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667c9468f0", "title": "Some Lessons on Reviews and Rebuttals", "url": "https://davidstutz.de/some-lessons-on-reviews-and-rebuttals/", "published_at": "2025-02-02T20:04:24+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667cc48c39", "title": "Thoughts on Watermarking AI-Generated Content", "url": "https://davidstutz.de/thoughts-on-watermarking-ai-generated-content/", "published_at": "2025-01-15T22:02:54+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667d9397d4", "title": "Thoughts and Lessons for Planning Rater Studies in AI", "url": "https://davidstutz.de/thoughts-and-lessons-for-planning-rater-studies-in-ai/", "published_at": "2025-01-06T13:54:12+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667e770b36", "title": "Open-Sourcing Relabeled MedQA and Dermatology DDx Datasets", "url": "https://davidstutz.de/open-sourcing-relabeled-medqa-and-dermatology-ddx-datasets/", "published_at": "2024-11-12T17:42:10+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667f2440cb", "title": "Thinking About Research Ideas vs. Technology", "url": "https://davidstutz.de/thinking-about-research-ideas-vs-technology/", "published_at": "2024-11-10T14:38:04+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667f5bcd78", "title": "The Importance of Effectively Experimenting in an AI PhD", "url": "https://davidstutz.de/the-importance-of-effectively-experimenting-in-an-ai-phd/", "published_at": "2024-07-08T12:16:28+00:00" }, { "id": "01a08769-d552-72d9-94c7-82667fa0f3ea", "title": "FAQ for our Monte Carlo Conformal Prediction", "url": "https://davidstutz.de/faq-for-our-monte-carlo-conformal-prediction/", "published_at": "2024-05-12T16:41:33+00:00" }, { "id": "01a0c506-a811-737d-a983-654014bd2dcb", "title": "Documenting your PhD — Keeping Track of Meetings, Experiments and Decisions", "url": "https://davidstutz.de/documenting-your-phd/", "published_at": "2024-05-06T16:53:04+00:00" }, { "id": "01a0c506-a811-737d-a983-6540156c9636", "title": "On NeurIPS’ High School Paper Track", "url": "https://davidstutz.de/on-neurips-high-school-paper-track/", "published_at": "2024-04-14T19:43:16+00:00" }, { "id": "01a0c506-a811-737d-a983-654016219ff2", "title": "Thoughts on Academia and Industry in Machine Learning Research", "url": "https://davidstutz.de/thoughts-on-academia-and-industry-in-machine-learning-research/", "published_at": "2024-03-27T20:22:34+00:00" }, { "id": "01a0c506-a811-737d-a983-6540166cd01b", "title": "On the Utility of Conformal Prediction Intervals", "url": "https://davidstutz.de/on-the-utility-of-conformal-prediction-intervals/", "published_at": "2024-02-18T14:19:31+00:00" }, { "id": "01a0c506-a811-737d-a983-6540167427ac", "title": "Vanderbilt Machine Learning Seminar Talk “Conformal Prediction under Ambiguous Ground Truth”", "url": "https://davidstutz.de/vanderbilt-machine-learning-seminar-talk-conformal-prediction-under-ambiguous-ground-truth/", "published_at": "2023-11-12T13:25:35+00:00" }, { "id": "01a0c506-a811-737d-a983-654016d0ee97", "title": "PRECISE Seminar Talk “Evaluating and Calibrating AI Models with Uncertain Ground Truth”", "url": "https://davidstutz.de/precise-seminar-talk-evaluating-and-calibrating-ai-models-with-uncertain-ground-truth/", "published_at": "2023-11-10T12:53:13+00:00" }, { "id": "01a0c506-a811-737d-a983-65401794c022", "title": "TMLR Paper “Conformal Prediction under Ambiguous Ground Truth”", "url": "https://davidstutz.de/tmlr-paper-conformal-prediction-under-ambiguous-ground-truth/", "published_at": "2023-11-02T23:48:55+00:00" }, { "id": "01a0c506-a811-737d-a983-6540187a168d", "title": "ArXiv Pre-Print “Evaluating AI Systems under Uncertain Ground Truth: a Case Study in Dermatology”", "url": "https://davidstutz.de/arxiv-pre-print-evaluating-ai-systems-under-uncertain-ground-truth-a-case-study-in-dermatology/", "published_at": "2023-11-01T13:06:48+00:00" }, { "id": "01a0c506-a811-737d-a983-65401890e01f", "title": "Interviewed by AI Coffee Break with Letitia", "url": "https://davidstutz.de/interviewed-by-ai-coffee-break-with-letitia/", "published_at": "2023-10-29T23:52:02+00:00" }, { "id": "01a0c506-a811-737d-a983-6540194d2ecd", "title": "Benchmarking Bit Errors in Quantized Neural Networks with PyTorch", "url": "https://davidstutz.de/benchmarking-bit-errors-in-quantized-neural-networks-with-pytorch/", "published_at": "2023-10-16T10:50:52+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f03763140dab", "title": "My Impressions (and Application) of the Heidelberg Laureate Forum 2023", "url": "https://davidstutz.de/my-impressions-and-application-of-the-heidelberg-laureate-forum-2023/", "published_at": "2023-10-04T20:31:05+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f03763fef16e", "title": "Awarded DAGM MVTec Dissertation Award 2023", "url": "https://davidstutz.de/awarded-dagm-mvtec-dissertation-award-2023/", "published_at": "2023-10-02T19:47:46+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f037643550a6", "title": "Simple Adversarial Transformations in PyTorch", "url": "https://davidstutz.de/simple-adversarial-transformations-in-pytorch/", "published_at": "2023-09-18T08:52:34+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f03764b52e54", "title": "Adversarial Patches and Frames in PyTorch", "url": "https://davidstutz.de/adversarial-patches-and-frames-in-pytorch/", "published_at": "2023-09-10T14:18:37+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f03765357142", "title": "Distal Adversarial Examples Against Neural Networks in PyTorch", "url": "https://davidstutz.de/distal-adversarial-examples-against-neural-networks-in-pytorch/", "published_at": "2023-09-05T17:36:55+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f037655505d1", "title": "Proper Robustness Evaluation of Confidence-Calibrated Adversarial Training in PyTorch", "url": "https://davidstutz.de/proper-robustness-evaluation-of-confidence-calibrated-adversarial-training-in-pytorch/", "published_at": "2023-07-20T16:48:52+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f03765e5ef9e", "title": "Guest on Jay Shah’s Machine Learning Podcast", "url": "https://davidstutz.de/guest-on-jay-shahs-machine-learning-podcast/", "published_at": "2023-07-01T19:36:32+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f037665a22a0", "title": "Generalizing Adversarial Robustness with Confidence-Calibrated Adversarial Training in PyTorch", "url": "https://davidstutz.de/generalizing-adversarial-robustness-with-confidence-calibrated-adversarial-training-in-pytorch/", "published_at": "2023-06-30T17:29:45+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f03766c1cf31", "title": "47.9% Robust Test Error on CIFAR10 with Adversarial Training and PyTorch", "url": "https://davidstutz.de/47-9-robust-test-error-on-cifar10-with-adversarial-training-and-pytorch/", "published_at": "2023-06-17T21:45:24+00:00" }, { "id": "01a0c517-e77f-702d-8d39-f03766f2b260", "title": "Some Research Ideas for Conformal Training", "url": "https://davidstutz.de/some-research-ideas-for-conformal-training/", "published_at": "2023-05-15T14:58:53+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa5d9e5d9c", "title": "Lp Adversarial Examples using Projected Gradient Descent in PyTorch", "url": "https://davidstutz.de/lp-adversarial-examples-using-projected-gradient-descent-in-pytorch/", "published_at": "2023-05-14T15:26:37+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa5e09ec79", "title": "2.56% Test Error on CIFAR-10 using PyTorch and AutoAugment", "url": "https://davidstutz.de/2-percent-test-error-on-cifar10-using-pytorch-autoagument/", "published_at": "2023-04-26T20:01:19+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa5e4bd7be", "title": "Loading and Saving PyTorch Models Without Knowing the Architecture in Advance", "url": "https://davidstutz.de/loading-and-saving-pytorch-models-without-knowing-the-architecture/", "published_at": "2023-04-26T20:01:13+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa5e59fa77", "title": "Monitoring PyTorch Training using Tensorboard", "url": "https://davidstutz.de/monitoring-training-in-pytorch-using-tensorboard/", "published_at": "2023-04-26T20:01:07+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa5f06fe2e", "title": "Updated Results for Confidence-Calibrated Adversarial Training", "url": "https://davidstutz.de/discussion-of-recent-results-for-confidence-calibrated-adversarial-training/", "published_at": "2023-04-04T15:36:41+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa5f92335d", "title": "Thoroughly Spell-Checking a PhD Thesis", "url": "https://davidstutz.de/thoroughly-spell-checking-a-phd-thesis/", "published_at": "2023-03-16T17:19:17+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa607da0fd", "title": "Python Scripts to Prepare ArXiv Submissions", "url": "https://davidstutz.de/python-scripts-to-prepare-arxiv-submissions/", "published_at": "2023-02-16T21:42:10+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa61562bc7", "title": "A PhD in Numbers", "url": "https://davidstutz.de/a-phd-in-numbers/", "published_at": "2023-01-08T21:53:10+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa61e4273e", "title": "What I Learned About PhD Programs — Updated 4 Years Later", "url": "https://davidstutz.de/what-i-learned-about-phd-programs-updated-4-years-later/", "published_at": "2022-11-18T18:09:37+00:00" }, { "id": "01a0c522-e2e4-7110-8ced-9dfa62a4e940", "title": "PhD Thesis on Robustness and Uncertainty in Deep Learning", "url": "https://davidstutz.de/phd-thesis-on-robustness-and-uncertainty-in-deep-learning/", "published_at": "2022-11-13T15:21:01+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba318ecba", "title": "PhD Defense Slides and Lessons Learned", "url": "https://davidstutz.de/phd-defense-slides-and-lessons-learned/", "published_at": "2022-10-27T22:16:02+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba4107063", "title": "How I Prepared for DeepMind and Google AI Research Internship Interviews in 2019", "url": "https://davidstutz.de/how-i-prepared-for-deepmind-and-google-ai-research-internship-interviews-in-2019/", "published_at": "2022-09-24T12:46:55+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba51066a1", "title": "Code Released: Conformal Training", "url": "https://davidstutz.de/code-released-conformal-training/", "published_at": "2022-08-17T16:33:26+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba52646be", "title": "ICML 2022 Art of Robustness Paper “On Fragile Features and Batch Normalization in Adversarial Training”", "url": "https://davidstutz.de/icml-2022-art-of-robustness-paper-on-fragile-features-and-batch-normalization-in-adversarial-training/", "published_at": "2022-08-05T14:38:42+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba5dad1b9", "title": "Machine Learning Security Seminar Talk “Relating Adversarially Robust Generalization to Flat Minima”", "url": "https://davidstutz.de/machine-learning-security-seminar-talk-relating-adversarially-robust-generalization-to-flat-minima/", "published_at": "2021-12-10T12:59:22+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba603c083", "title": "International Seminar on Distribution-Free Statistics Talk “Conformal Training: Learning Optimal Conformal Classifiers”", "url": "https://davidstutz.de/international-seminar-on-distribution-free-statistics-talk-conformal-training-learning-optimal-conformal-classifiers/", "published_at": "2021-11-22T09:10:29+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba68c29ce", "title": "Code Released: Adversarial Robust Generalization and Flatness", "url": "https://davidstutz.de/code-released-adversarial-robust-generalization-and-flatness/", "published_at": "2021-10-27T16:38:22+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba7163974", "title": "Math Machine Learning Seminar of MPI MiS and UCLA Talk “Relating Adversarial Robustness and Weight Robustness Through Flatness”", "url": "https://davidstutz.de/math-machine-learning-seminar-of-mpi-mis-and-ucla-talk-relating-adversarial-robustness-and-weight-robustness-through-flatness/", "published_at": "2021-10-26T14:28:56+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba7d0fece", "title": "ArXiv Pre-Print “Learning Optimal Conformal Classifiers”", "url": "https://davidstutz.de/arxiv-pre-print-learning-optimal-conformal-classifiers/", "published_at": "2021-10-21T11:27:37+00:00" }, { "id": "01a0c52e-6822-73d6-9fee-906ba8a435c4", "title": "Recorded ICCV’21 Talk “Relating Adversarially Robust Generalization to Flat Minima”", "url": "https://davidstutz.de/recorded-iccv21-talk-relating-adversarially-robust-generalization-to-flat-minima/", "published_at": "2021-10-12T12:27:40+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4e6d708c6", "title": "Qualcomm Innovation Fellowship Talk “Confidence-Calibrated Adversarial Training and Random Bit Error Training”", "url": "https://davidstutz.de/qualcomm-innovation-fellowship-talk-confidence-calibrated-adversarial-training-and-random-bit-error-training/", "published_at": "2021-07-22T11:01:03+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4e7bd6e32", "title": "Recorded CVPR’21 CV-AML Workshop Outstanding Paper Talk “Bit Error Robustness for Energy-Efficient DNN Accelerators”", "url": "https://davidstutz.de/recorded-cvpr21-cv-aml-workshop-outstanding-paper-talk-bit-error-robustness-for-energy-efficient-dnn-accelerators/", "published_at": "2021-07-03T11:32:13+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4e7d1c33b", "title": "ArXiv Pre-Print “Random and Adversarial Bit Error Robustness: Energy-Efficient and Secure DNN Accelerators”", "url": "https://davidstutz.de/arxiv-pre-print-random-and-adversarial-bit-error-robustness-energy-efficient-and-secure-dnn-accelerators/", "published_at": "2021-06-03T13:58:13+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4e8724ba0", "title": "ArXiv Pre-Print “Relating Adversarially Robust Generalization to Flat Minima”", "url": "https://davidstutz.de/arxiv-pre-print-relating-adversarially-robust-generalization-to-flat-minima/", "published_at": "2021-05-03T13:40:06+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4e8b92677", "title": "Recorded MLSys’21 Talk “Bit Error Robustness for Energy-Efficient DNN Accelerators”", "url": "https://davidstutz.de/recorded-mlsys21-talk-bit-error-robustness-for-energy-efficient-dnn-accelerators/", "published_at": "2021-05-01T16:46:25+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4e976430e", "title": "Talk at TU Dortmund “Random and Adversarial Bit Error Robustness of DNNs”", "url": "https://davidstutz.de/talk-at-tu-dortmund-random-and-adversarial-bit-error-robustness-of-dnns/", "published_at": "2021-04-30T10:54:36+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4e98e1efb", "title": "Recorded RobustAI Workshop Talk “Confidence-Calibrated Adversarial Training and Bit Error Robustness of DNNs”", "url": "https://davidstutz.de/recorded-robustai-workshop-talk-confidence-calibrated-adversarial-training-and-bit-error-robustness-of-dnns/", "published_at": "2021-01-19T13:48:43+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4e9a2cf64", "title": "Recorded FOCA’20 Talk “Bit Error Robustness for Energy-Efficient DNN Accelerators”", "url": "https://davidstutz.de/recorded-foca20-talk-bit-error-robustness-for-energy-efficient-dnn-accelerators/", "published_at": "2021-01-18T16:46:18+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4ea3c0808", "title": "Recorded ICML’20 Talk “Confidence-Calibrated Adversarial Training”", "url": "https://davidstutz.de/recorded-icml20-talk-confidence-calibrated-adversarial-training/", "published_at": "2021-01-11T12:24:43+00:00" }, { "id": "01a0c538-ec68-723a-8a8f-35e4eb31d542", "title": "Updated Pre-Print “Bit Error Robustness for Energy-Efficient DNN Accelerators “", "url": "https://davidstutz.de/updated-pre-print-bit-error-robustness-for-energy-efficient-dnn-accelerators/", "published_at": "2020-11-09T17:12:03+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b5284ca76b75", "title": "Code Released: Adversarial Patch Training", "url": "https://davidstutz.de/code-released-adversarial-patch-training/", "published_at": "2020-08-04T09:08:37+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b5284da5885f", "title": "What Lp Adversarial Examples make Sense on Common Vision Datasets?", "url": "https://davidstutz.de/what-lp-adversarial-examples-make-sense-on-common-vision-datasets/", "published_at": "2020-07-21T16:18:40+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b5284dca0d9a", "title": "ICML Talk “Confidence-Calibrated Adversarial Training”", "url": "https://davidstutz.de/icml-talk-confidence-calibrated-adversarial-training/", "published_at": "2020-07-03T11:27:55+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b5284e7707d4", "title": "ICML Paper “Confidence-Calibrated Adversarial Training”", "url": "https://davidstutz.de/icml-paper-confidence-calibrated-adversarial-training/", "published_at": "2020-07-01T09:05:43+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b5284eda7889", "title": "ArXiv Pre-Print “On Mitigating Random and Adversarial Bit Errors”", "url": "https://davidstutz.de/arxiv-pre-print-on-mitigating-random-and-adversarial-bit-errors/", "published_at": "2020-06-26T12:43:52+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b5284fc405a9", "title": "“Cross-Entropy Loss Leads To Poor Margins”, Nar et al.", "url": "https://davidstutz.de/cross-entropy-loss-leads-to-poor-margins-nar-et-al/", "published_at": "2020-06-25T15:30:58+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b5285020a92e", "title": "“Regularizing by the Variance of the Activations’ Sample-Variances”, Littwin and Wolf", "url": "https://davidstutz.de/regularizing-by-the-variance-of-the-activations-sample-variances-littwin-and-wolf/", "published_at": "2020-06-23T15:30:56+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b52850e6fec9", "title": "“A Spectral View of Adversarially Robust Features”, Garg et al.", "url": "https://davidstutz.de/a-spectral-view-of-adversarially-robust-features-garg-et-al/", "published_at": "2020-06-18T15:30:12+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b52851d11b68", "title": "“Adversarial Examples Are a Natural Consequence of Test Error in Noise”, Ford et al.", "url": "https://davidstutz.de/adversarial-examples-are-a-natural-consequence-of-test-error-in-noise-ford-et-al/", "published_at": "2020-06-16T15:30:08+00:00" }, { "id": "01a0c542-5081-70e0-92eb-b5285290edf7", "title": "“Sharp Minima Can Generalize For Deep Nets”, Dinh et al.", "url": "https://davidstutz.de/sharp-minima-can-generalize-for-deep-nets-dinh-et-al/", "published_at": "2020-06-11T15:29:30+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f975da4240", "title": "“On Correlation of Features Extracted by Deep Neural Networks”, Ayinde et al.", "url": "https://davidstutz.de/on-correlation-of-features-extracted-by-deep-neural-networks-ayinde-et-al/", "published_at": "2020-06-09T15:29:29+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f975ead3d1", "title": "“A Research Agenda: Dynamic Models to Defend Against Correlated Attacks”, Goodfellow", "url": "https://davidstutz.de/a-research-agenda-dynamic-models-to-defend-against-correlated-attacks-goodfellow/", "published_at": "2020-06-04T15:27:56+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f976ae6feb", "title": "Illustrating (Convolutional) Neural Networks in LaTeX with TikZ", "url": "https://davidstutz.de/illustrating-convolutional-neural-networks-in-latex-with-tikz/", "published_at": "2020-06-02T19:14:36+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f977027038", "title": "“Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels”, Zhang et al.", "url": "https://davidstutz.de/generalized-cross-entropy-loss-for-training-deep-neural-networks-with-noisy-labels-zhang-et-al/", "published_at": "2020-06-02T15:27:53+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f977054386", "title": "“Group Normalization”, Wu and He", "url": "https://davidstutz.de/group-normalization-wu-and-he/", "published_at": "2020-05-28T15:27:06+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f9770d0871", "title": "“Instance Normalization: The Missing Ingredient for Fast Stylization”, Ulyanov and Vedaldi", "url": "https://davidstutz.de/instance-normalization-the-missing-ingredient-for-fast-stylization-ulyanov-and-vedaldi/", "published_at": "2020-05-26T15:27:04+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f9772cec48", "title": "“Sensitivity and Generalization in Neural Networks: an Empirical Study”, Novak et al.", "url": "https://davidstutz.de/sensitivity-and-generalization-in-neural-networks-an-empirical-study-novak-et-al/", "published_at": "2020-05-21T15:26:28+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f9782356c2", "title": "“Layer Normalization”, Ba et al.", "url": "https://davidstutz.de/layer-normalization-ba-et-al-2/", "published_at": "2020-05-19T15:26:25+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f9791d162b", "title": "“Bayesian Uncertainty Estimation for Batch Normalized Deep Networks”, Teye et al.", "url": "https://davidstutz.de/bayesian-uncertainty-estimation-for-batch-normalized-deep-networks-teye-et-al/", "published_at": "2020-05-14T15:25:39+00:00" }, { "id": "01a0c54b-2896-723a-b2ce-01f97989b9a7", "title": "“MNIST-C: A Robustness Benchmark for Computer Vision”, Mu and Gilmer", "url": "https://davidstutz.de/mnist-c-a-robustness-benchmark-for-computer-vision-mu-and-gilmer/", "published_at": "2020-05-12T15:25:38+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5ac28a03", "title": "ArXiv Pre-Print “Adversarial Training against Location-Optimized Adversarial Patches”", "url": "https://davidstutz.de/arxiv-pre-print-adversarial-training-against-location-optimized-adversarial-patches/", "published_at": "2020-05-08T16:05:50+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5b53795f", "title": "“Benchmarking Neural Network Robustness to Common Corruptions and Perturbations”, Hendrycks and Dietterich", "url": "https://davidstutz.de/benchmarking-neural-network-robustness-to-common-corruptions-and-perturbations-hendrycks-and-dietterich/", "published_at": "2020-05-07T15:24:47+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5c0bc587", "title": "“Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation”, Lopes et al.", "url": "https://davidstutz.de/improving-robustness-without-sacrificing-accuracy-with-patch-gaussian-augmentation-lopes-et-al/", "published_at": "2020-05-05T15:24:49+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5cd9a399", "title": "Implementing Custom PyTorch Tensor Operations in C and CUDA", "url": "https://davidstutz.de/implementing-custom-pytorch-tensor-operations-in-c-and-cuda/", "published_at": "2020-04-30T16:09:17+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5daaad44", "title": "“The Space of Transferable Adversarial Examples”, Tramer et al.", "url": "https://davidstutz.de/the-space-of-transferable-adversarial-examples-tramer-et-al/", "published_at": "2020-04-30T15:23:45+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5e0544dd", "title": "“Efficient Evaluation-Time Uncertainty Estimation by Improved Distillation”, Englesson and Azizpour", "url": "https://davidstutz.de/efficient-evaluation-time-uncertainty-estimation-by-improved-distillation-englesson-and-azizpour/", "published_at": "2020-04-28T15:23:47+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5e80b815", "title": "“CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks”, Marchisio", "url": "https://davidstutz.de/capsattacks-robust-and-imperceptible-adversarial-attacks-on-capsule-networks-marchisio/", "published_at": "2020-04-23T15:23:11+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5f4dd027", "title": "“Exploring the Hyperparameter Landscape of Adversarial Robustness”, Düsterwald et al.", "url": "https://davidstutz.de/exploring-the-hyperparameter-landscape-of-adversarial-robustness-dusterwald-et-al/", "published_at": "2020-04-21T15:22:51+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f5fbeff81", "title": "Adversarial Training Has Higher Sample Complexity", "url": "https://davidstutz.de/adversarial-training-has-higher-sample-complexity/", "published_at": "2020-04-17T14:17:24+00:00" }, { "id": "01a0c553-70fb-7342-b17b-699f606ce333", "title": "“For Valid Generalization the Size of the Weights is More Important than the Size of the Network”, Barlett", "url": "https://davidstutz.de/for-valid-generalization-the-size-of-the-weights-is-more-important-than-the-size-of-the-network-barlett/", "published_at": "2020-04-16T15:21:59+00:00" }, { "id": "01a0c55b-dd57-710b-b32a-6d607a44a6db", "title": "“Interpolated Adversarial Training: Achieving Robust Neural Networks Without Sacrificing Too Much Accuracy”, Lamb et al.", "url": "https://davidstutz.de/interpolated-adversarial-training-achieving-robust-neural-networks-without-sacrificing-too-much-accuracy-lamb-et-al/", "published_at": "2020-04-14T15:21:26+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c51e5fe06", "title": "“Provably Robust Deep Learning via Adversarially Trained Smoothed Classifiers”, Salman et al.", "url": "https://davidstutz.de/provably-robust-deep-learning-via-adversarially-trained-smoothed-classifiers-salman-et-al/", "published_at": "2020-04-10T15:20:52+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c52012682", "title": "“DPATCH: An Adversarial Patch Attack on Object Detectors”, Liu et al.", "url": "https://davidstutz.de/dpatch-an-adversarial-patch-attack-on-object-detectors-liu-et-al/", "published_at": "2020-04-06T15:20:09+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c528d7b38", "title": "“Towards Robust, Locally Linear Deep Networks”, Lee et al.", "url": "https://davidstutz.de/towards-robust-locally-linear-deep-networks-lee-et-al/", "published_at": "2020-04-03T15:19:43+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c52eb908b", "title": "“Exploiting the Inherent Limitation of L0 Adversarial Examples”, Zuo et al.", "url": "https://davidstutz.de/exploiting-the-inherent-limitation-of-l0-adversarial-examples-zuo-et-al/", "published_at": "2020-04-01T15:19:20+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c52fd0274", "title": "On-Manifold Adversarial Training for Boosting Generalization", "url": "https://davidstutz.de/on-manifold-adversarial-training-for-boosting-generalization/", "published_at": "2020-03-30T21:24:23+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c53d30857", "title": "“LaVAN: Localized and Visible Adversarial Noise”, Karmon et al.", "url": "https://davidstutz.de/lavan-localized-and-visible-adversarial-noise-karmon-et-al/", "published_at": "2020-03-30T15:18:50+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c53df84e4", "title": "“Semantic Adversarial Examples”, Hosseini and Poovendran", "url": "https://davidstutz.de/semantic-adversarial-examples-hosseini-and-poovendran/", "published_at": "2020-03-27T16:18:30+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c54c5cf22", "title": "“Low Frequency Adversarial Perturbation”, Guo et al.", "url": "https://davidstutz.de/low-frequency-adversarial-perturbation-guo-et-al/", "published_at": "2020-03-25T16:18:05+00:00" }, { "id": "01a0c55b-dd5c-732e-a956-8f2c552a5f61", "title": "“Thwarting Adversarial Examples: An L_0-Robust Sparse Fourier Transform”, Bafna et al.", "url": "https://davidstutz.de/thwarting-adversarial-examples-an-l_0-robust-sparse-fourier-transform-bafna-et-al/", "published_at": "2020-03-23T16:17:34+00:00" } ] posts Claim your blog
Back to davidstutz.de
Blog · corpus.blog/blogs/davidstutz.de/posts

davidstutz.de

davidstutz.de

2026

Domain-Specific AI Should Focus on Workflows Rather Than Modelingoriginal ↗
14 Jun 2026
AI Evaluation is Becoming an Exciting Standalone Disciplineoriginal ↗
16 Apr 2026

2025

RAISE 2025 Panel Statement on Aligning AI to Clinical Valuesoriginal ↗
4 Oct 2025
Some Lessons on Reviews and Rebuttalsoriginal ↗
2 Feb 2025
Thoughts on Watermarking AI-Generated Contentoriginal ↗
15 Jan 2025
Thoughts and Lessons for Planning Rater Studies in AIoriginal ↗
6 Jan 2025

2024

Open-Sourcing Relabeled MedQA and Dermatology DDx Datasetsoriginal ↗
12 Nov 2024
Thinking About Research Ideas vs. Technologyoriginal ↗
10 Nov 2024
The Importance of Effectively Experimenting in an AI PhDoriginal ↗
8 Jul 2024
FAQ for our Monte Carlo Conformal Predictionoriginal ↗
12 May 2024
Documenting your PhD — Keeping Track of Meetings, Experiments and Decisionsoriginal ↗
6 May 2024
On NeurIPS’ High School Paper Trackoriginal ↗
14 Apr 2024
Thoughts on Academia and Industry in Machine Learning Researchoriginal ↗
27 Mar 2024
On the Utility of Conformal Prediction Intervalsoriginal ↗
18 Feb 2024

2023

Vanderbilt Machine Learning Seminar Talk “Conformal Prediction under Ambiguous Ground Truth”original ↗
12 Nov 2023
PRECISE Seminar Talk “Evaluating and Calibrating AI Models with Uncertain Ground Truth”original ↗
10 Nov 2023
TMLR Paper “Conformal Prediction under Ambiguous Ground Truth”original ↗
2 Nov 2023
ArXiv Pre-Print “Evaluating AI Systems under Uncertain Ground Truth: a Case Study in Dermatology”original ↗
1 Nov 2023
Interviewed by AI Coffee Break with Letitiaoriginal ↗
29 Oct 2023
Benchmarking Bit Errors in Quantized Neural Networks with PyTorchoriginal ↗
16 Oct 2023
My Impressions (and Application) of the Heidelberg Laureate Forum 2023original ↗
4 Oct 2023
Awarded DAGM MVTec Dissertation Award 2023original ↗
2 Oct 2023
Simple Adversarial Transformations in PyTorchoriginal ↗
18 Sept 2023
Adversarial Patches and Frames in PyTorchoriginal ↗
10 Sept 2023
Distal Adversarial Examples Against Neural Networks in PyTorchoriginal ↗
5 Sept 2023
Proper Robustness Evaluation of Confidence-Calibrated Adversarial Training in PyTorchoriginal ↗
20 Jul 2023
Guest on Jay Shah’s Machine Learning Podcastoriginal ↗
1 Jul 2023
Generalizing Adversarial Robustness with Confidence-Calibrated Adversarial Training in PyTorchoriginal ↗
30 Jun 2023
47.9% Robust Test Error on CIFAR10 with Adversarial Training and PyTorchoriginal ↗
17 Jun 2023
Some Research Ideas for Conformal Trainingoriginal ↗
15 May 2023
Lp Adversarial Examples using Projected Gradient Descent in PyTorchoriginal ↗
14 May 2023
2.56% Test Error on CIFAR-10 using PyTorch and AutoAugmentoriginal ↗
26 Apr 2023
Loading and Saving PyTorch Models Without Knowing the Architecture in Advanceoriginal ↗
26 Apr 2023
Monitoring PyTorch Training using Tensorboardoriginal ↗
26 Apr 2023
Updated Results for Confidence-Calibrated Adversarial Trainingoriginal ↗
4 Apr 2023
Thoroughly Spell-Checking a PhD Thesisoriginal ↗
16 Mar 2023
Python Scripts to Prepare ArXiv Submissionsoriginal ↗
16 Feb 2023
A PhD in Numbersoriginal ↗
8 Jan 2023

2022

What I Learned About PhD Programs — Updated 4 Years Lateroriginal ↗
18 Nov 2022
PhD Thesis on Robustness and Uncertainty in Deep Learningoriginal ↗
13 Nov 2022
PhD Defense Slides and Lessons Learnedoriginal ↗
27 Oct 2022
How I Prepared for DeepMind and Google AI Research Internship Interviews in 2019original ↗
24 Sept 2022
Code Released: Conformal Trainingoriginal ↗
17 Aug 2022
ICML 2022 Art of Robustness Paper “On Fragile Features and Batch Normalization in Adversarial Training”original ↗
5 Aug 2022

2021

Machine Learning Security Seminar Talk “Relating Adversarially Robust Generalization to Flat Minima”original ↗
10 Dec 2021
International Seminar on Distribution-Free Statistics Talk “Conformal Training: Learning Optimal Conformal Classifiers”original ↗
22 Nov 2021
Code Released: Adversarial Robust Generalization and Flatnessoriginal ↗
27 Oct 2021
Math Machine Learning Seminar of MPI MiS and UCLA Talk “Relating Adversarial Robustness and Weight Robustness Through Flatness”original ↗
26 Oct 2021
ArXiv Pre-Print “Learning Optimal Conformal Classifiers”original ↗
21 Oct 2021
Recorded ICCV’21 Talk “Relating Adversarially Robust Generalization to Flat Minima”original ↗
12 Oct 2021
Qualcomm Innovation Fellowship Talk “Confidence-Calibrated Adversarial Training and Random Bit Error Training”original ↗
22 Jul 2021
Recorded CVPR’21 CV-AML Workshop Outstanding Paper Talk “Bit Error Robustness for Energy-Efficient DNN Accelerators”original ↗
3 Jul 2021
ArXiv Pre-Print “Random and Adversarial Bit Error Robustness: Energy-Efficient and Secure DNN Accelerators”original ↗
3 Jun 2021
ArXiv Pre-Print “Relating Adversarially Robust Generalization to Flat Minima”original ↗
3 May 2021
Recorded MLSys’21 Talk “Bit Error Robustness for Energy-Efficient DNN Accelerators”original ↗
1 May 2021
Talk at TU Dortmund “Random and Adversarial Bit Error Robustness of DNNs”original ↗
30 Apr 2021
Recorded RobustAI Workshop Talk “Confidence-Calibrated Adversarial Training and Bit Error Robustness of DNNs”original ↗
19 Jan 2021
Recorded FOCA’20 Talk “Bit Error Robustness for Energy-Efficient DNN Accelerators”original ↗
18 Jan 2021
Recorded ICML’20 Talk “Confidence-Calibrated Adversarial Training”original ↗
11 Jan 2021

2020

Updated Pre-Print “Bit Error Robustness for Energy-Efficient DNN Accelerators “original ↗
9 Nov 2020
Code Released: Adversarial Patch Trainingoriginal ↗
4 Aug 2020
What Lp Adversarial Examples make Sense on Common Vision Datasets?original ↗
21 Jul 2020
ICML Talk “Confidence-Calibrated Adversarial Training”original ↗
3 Jul 2020
ICML Paper “Confidence-Calibrated Adversarial Training”original ↗
1 Jul 2020
ArXiv Pre-Print “On Mitigating Random and Adversarial Bit Errors”original ↗
26 Jun 2020
“Cross-Entropy Loss Leads To Poor Margins”, Nar et al.original ↗
25 Jun 2020
“Regularizing by the Variance of the Activations’ Sample-Variances”, Littwin and Wolforiginal ↗
23 Jun 2020
“A Spectral View of Adversarially Robust Features”, Garg et al.original ↗
18 Jun 2020
“Adversarial Examples Are a Natural Consequence of Test Error in Noise”, Ford et al.original ↗
16 Jun 2020
“Sharp Minima Can Generalize For Deep Nets”, Dinh et al.original ↗
11 Jun 2020
“On Correlation of Features Extracted by Deep Neural Networks”, Ayinde et al.original ↗
9 Jun 2020
“A Research Agenda: Dynamic Models to Defend Against Correlated Attacks”, Goodfelloworiginal ↗
4 Jun 2020
Illustrating (Convolutional) Neural Networks in LaTeX with TikZoriginal ↗
2 Jun 2020
“Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels”, Zhang et al.original ↗
2 Jun 2020
“Group Normalization”, Wu and Heoriginal ↗
28 May 2020
“Instance Normalization: The Missing Ingredient for Fast Stylization”, Ulyanov and Vedaldioriginal ↗
26 May 2020
“Sensitivity and Generalization in Neural Networks: an Empirical Study”, Novak et al.original ↗
21 May 2020
“Layer Normalization”, Ba et al.original ↗
19 May 2020
“Bayesian Uncertainty Estimation for Batch Normalized Deep Networks”, Teye et al.original ↗
14 May 2020
“MNIST-C: A Robustness Benchmark for Computer Vision”, Mu and Gilmeroriginal ↗
12 May 2020
ArXiv Pre-Print “Adversarial Training against Location-Optimized Adversarial Patches”original ↗
8 May 2020
“Benchmarking Neural Network Robustness to Common Corruptions and Perturbations”, Hendrycks and Dietterichoriginal ↗
7 May 2020
“Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation”, Lopes et al.original ↗
5 May 2020
Implementing Custom PyTorch Tensor Operations in C and CUDAoriginal ↗
30 Apr 2020
“The Space of Transferable Adversarial Examples”, Tramer et al.original ↗
30 Apr 2020
“Efficient Evaluation-Time Uncertainty Estimation by Improved Distillation”, Englesson and Azizpouroriginal ↗
28 Apr 2020
“CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks”, Marchisiooriginal ↗
23 Apr 2020
“Exploring the Hyperparameter Landscape of Adversarial Robustness”, Düsterwald et al.original ↗
21 Apr 2020
Adversarial Training Has Higher Sample Complexityoriginal ↗
17 Apr 2020
“For Valid Generalization the Size of the Weights is More Important than the Size of the Network”, Barlettoriginal ↗
16 Apr 2020
“Interpolated Adversarial Training: Achieving Robust Neural Networks Without Sacrificing Too Much Accuracy”, Lamb et al.original ↗
14 Apr 2020
“Provably Robust Deep Learning via Adversarially Trained Smoothed Classifiers”, Salman et al.original ↗
10 Apr 2020
“DPATCH: An Adversarial Patch Attack on Object Detectors”, Liu et al.original ↗
6 Apr 2020
“Towards Robust, Locally Linear Deep Networks”, Lee et al.original ↗
3 Apr 2020
“Exploiting the Inherent Limitation of L0 Adversarial Examples”, Zuo et al.original ↗
1 Apr 2020
On-Manifold Adversarial Training for Boosting Generalizationoriginal ↗
30 Mar 2020
“LaVAN: Localized and Visible Adversarial Noise”, Karmon et al.original ↗
30 Mar 2020
“Semantic Adversarial Examples”, Hosseini and Poovendranoriginal ↗
27 Mar 2020
“Low Frequency Adversarial Perturbation”, Guo et al.original ↗
25 Mar 2020
“Thwarting Adversarial Examples: An L_0-Robust Sparse Fourier Transform”, Bafna et al.original ↗
23 Mar 2020
Next