corpus.blog
41554 blogs · [ { "id": "01a0c564-32ca-70c5-bcf0-953405b41bf8", "title": "“Adversarial camera stickers: A physical camera-based attack on deep learning systems”, Li et al.", "url": "https://davidstutz.de/adversarial-camera-stickers-a-physical-camera-based-attack-on-deep-learning-systems-li-et-al/", "published_at": "2020-03-23T16:17:10+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-9534066e769b", "title": "“On Norm-Agnostic Robustness of Adversarial Training”, Li et al.", "url": "https://davidstutz.de/on-norm-agnostic-robustness-of-adversarial-training-li-et-al/", "published_at": "2020-03-20T16:16:25+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-953406d7937a", "title": "“Adversarially Robust Distillation”, Goldblum et al.", "url": "https://davidstutz.de/adversarially-robust-distillation-goldblum-et-al/", "published_at": "2020-03-19T16:15:58+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-9534072e23c3", "title": "“Local Gradients Smoothing: Defense Against Localized Adversarial Attacks”, Naseer et al.", "url": "https://davidstutz.de/local-gradients-smoothing-defense-against-localized-adversarial-attacks-naseer-et-al/", "published_at": "2020-03-17T16:15:25+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-953407f8ab90", "title": "Adversarial Examples Leave the Data Manifold", "url": "https://davidstutz.de/adversarial-examples-leave-the-data-manifold/", "published_at": "2020-03-12T16:20:50+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-953408b620c9", "title": "Code Released: Confidence-Calibrated Adversarial Training", "url": "https://davidstutz.de/code-released-confidence-calibrated-adversarial-training/", "published_at": "2020-03-01T14:22:03+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-953408babbc6", "title": "Talk on Confidence-Calibrated Adversarial Training at BCAI and Tübingen AI Center", "url": "https://davidstutz.de/talk-on-confidence-calibrated-adversarial-training-at-bcai-and-tubingen-ai-center/", "published_at": "2020-02-28T13:21:43+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-953409a94c7a", "title": "Updated ArXiv Pre-Print “Confidence-Calibrated Adversarial Training”", "url": "https://davidstutz.de/updated-arxiv-pre-print-confidence-calibrated-adversarial-training/", "published_at": "2020-02-26T19:29:23+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-95340a5aef4b", "title": "On-Manifold Adversarial Examples", "url": "https://davidstutz.de/on-manifold-adversarial-examples/", "published_at": "2020-02-11T17:24:04+00:00" }, { "id": "01a0c564-32ca-70c5-bcf0-95340b263f89", "title": "FONTS: A Synthetic MNIST-Like Dataset with Known Manifold", "url": "https://davidstutz.de/fonts-a-synthetic-mnist-like-dataset-with-known-manifold/", "published_at": "2020-01-08T14:49:46+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7accecd99cb", "title": "240+ Papers on Adversarial Examples and Out-of-Distribution Detection", "url": "https://davidstutz.de/240-papers-on-adversarial-examples-and-out-of-distribution-detection/", "published_at": "2019-12-10T08:43:33+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7acced92380", "title": "A Short Introduction to Bayesian Neural Networks", "url": "https://davidstutz.de/a-short-introduction-to-bayesian-neural-networks/", "published_at": "2019-11-25T15:38:51+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7accf32cbc0", "title": "AI and Deep Learning at the 7th Heidelberg Laureate Forum 2019", "url": "https://davidstutz.de/ai-and-deep-learning-at-the-7th-heidelberg-laureate-forum-2019/", "published_at": "2019-11-06T13:00:54+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7accf912103", "title": "“Robustness and generalization”, Xu and Mannor", "url": "https://davidstutz.de/robustness-and-generalization-xu-and-mannor/", "published_at": "2019-11-02T18:20:18+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7acd0198a55", "title": "“Second-Order Adversarial Attack and Certifiable Robustness”, Li et al.", "url": "https://davidstutz.de/second-order-adversarial-attack-and-certifiable-robustness-li-et-al/", "published_at": "2019-11-01T17:54:15+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7acd090ebbb", "title": "“Certified Robustness to Adversarial Examples with Differential Privacy”, Lecuyer et al.", "url": "https://davidstutz.de/certified-robustness-to-adversarial-examples-with-differential-privacy-lecuyer-et-al/", "published_at": "2019-10-30T17:50:50+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7acd0b59bd7", "title": "More Examples for Working with Torch", "url": "https://davidstutz.de/more-examples-for-working-with-torch/", "published_at": "2019-10-29T18:13:12+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7acd1328a02", "title": "“ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness”, Geirhos et al.", "url": "https://davidstutz.de/imagenet-trained-cnns-are-biased-towards-texture-increasing-shape-bias-improves-accuracy-and-robustness-geirhos-et-al/", "published_at": "2019-10-29T17:36:32+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7acd1db475e", "title": "“An Empirical Evaluation on Robustness and Uncertainty of Regularization Methods”, Chun et al.", "url": "https://davidstutz.de/an-empirical-evaluation-on-robustness-and-uncertainty-of-regularization-methods-chun-et-al/", "published_at": "2019-10-28T17:15:59+00:00" }, { "id": "01a0c56b-6a83-7044-8647-f7acd265f6d0", "title": "“Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet”, Brendel and Bethge", "url": "https://davidstutz.de/approximating-cnns-with-bag-of-local-features-models-works-surprisingly-well-on-imagenet-brendel-and-bethge/", "published_at": "2019-10-26T16:08:36+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb7615f6f4f9", "title": "“Towards Stable and Efficient Training of Verifiably Robust Neural Networks”, Zhang et al.", "url": "https://davidstutz.de/towards-stable-and-efficient-training-of-verifiably-robust-neural-networks-zhang-et-al/", "published_at": "2019-10-24T15:57:51+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb7616969224", "title": "“Efficient Neural Network Robustness Certification with General Activation Functions”, Zhang et al.", "url": "https://davidstutz.de/efficient-neural-network-robustness-certification-with-general-activation-functions-zhang-et-al/", "published_at": "2019-10-23T15:53:53+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb7617601ce3", "title": "“Generalization in Deep Networks: The Role of Distance from Initialization”, Nagarajan and Kolter", "url": "https://davidstutz.de/generalization-in-deep-networks-the-role-of-distance-from-initialization-nagarajan-and-kolter/", "published_at": "2019-10-22T15:49:33+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb7617ed675f", "title": "“Likelihood Ratios for Out-of-Distribution Detection”, Ren et al.", "url": "https://davidstutz.de/likelihood-ratios-for-out-of-distribution-detection-ren-et-al/", "published_at": "2019-10-21T17:53:03+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb7618c2ea99", "title": "“On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Models”, Gowal et al.", "url": "https://davidstutz.de/on-the-effectiveness-of-interval-bound-propagation-for-training-verifiably-robust-models-gowal-et-al/", "published_at": "2019-10-21T15:44:27+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb7619361434", "title": "“Batch Normalization is a Cause of Adversarial Vulnerability”, Galloway et al.", "url": "https://davidstutz.de/batch-normalization-is-a-cause-of-adversarial-vulnerability-galloway-et-al/", "published_at": "2019-10-20T15:42:06+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb7619b62c60", "title": "“Detecting Extrapolation with Influence Functions”, Madras et al.", "url": "https://davidstutz.de/detecting-extrapolation-with-influence-functions-madras-et-al/", "published_at": "2019-10-19T17:48:32+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb7619f34ec8", "title": "“Radial basis function neural networks: a topical state-of-the-artsurvey”, Dash et al.", "url": "https://davidstutz.de/radial-basis-function-neural-networks-a-topical-state-of-the-artsurvey-dash-et-al/", "published_at": "2019-10-18T17:41:10+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb761ac86d7b", "title": "“How Can We Be So Dense? The Benefits of Using Highly Sparse Representations”, Ahmad and Scheinkman", "url": "https://davidstutz.de/how-can-we-be-so-dense-the-benefits-of-using-highly-sparse-representations-ahmad-and-scheinkman/", "published_at": "2019-10-17T17:30:41+00:00" }, { "id": "01a0c574-ee75-71af-be14-cb761b2c3356", "title": "“Deep-RBF Networks Revisited: Robust Classification with Rejection”, Zadeh et al.", "url": "https://davidstutz.de/deep-rbf-networks-revisited-robust-classification-with-rejection-zadeh-et-al/", "published_at": "2019-10-16T17:23:22+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6999863d", "title": "ArXiv Pre-Print “Confidence-Calibrated Adversarial Training”", "url": "https://davidstutz.de/arxiv-pre-print-confidence-calibrated-adversarial-training/", "published_at": "2019-10-16T16:54:02+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6a176d8a", "title": "“Neural Networks with Structural Resistance to Adversarial Attacks”, De ALfaro", "url": "https://davidstutz.de/neural-networks-with-structural-resistance-to-adversarial-attacks-de-alfaro/", "published_at": "2019-10-15T17:21:07+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6a8d9dc0", "title": "“Adversarial Examples Are Not Bugs, They Are Features”, Ilyas et al.", "url": "https://davidstutz.de/adversarial-examples-are-not-bugs-they-are-features-ilyas-et-al/", "published_at": "2019-10-13T17:10:37+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6aa82644", "title": "“Bit-Flip Attack: Crushing Neural Network withProgressive Bit Search”, Rakin et al.", "url": "https://davidstutz.de/bit-flip-attack-crushing-neural-network-withprogressive-bit-search-rakin-et-al/", "published_at": "2019-10-12T17:05:14+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6ac76c8f", "title": "“The Lottery Ticket Hypothesis: Training Pruned Neural Networks”, Frankle and Carbin", "url": "https://davidstutz.de/the-lottery-ticket-hypothesis-training-pruned-neural-networks-frankle-and-carbin/", "published_at": "2019-10-11T19:58:42+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6b9eecc9", "title": "“Certified Adversarial Robustness via Randomized Smoothing”, Cohen et al.", "url": "https://davidstutz.de/certified-adversarial-robustness-via-randomized-smoothing-cohen-et-al/", "published_at": "2019-10-08T19:41:58+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6bc3a33a", "title": "“Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks”, Liang et al.", "url": "https://davidstutz.de/enhancing-the-reliability-of-out-of-distribution-image-detection-in-neural-networks-liang-et-al/", "published_at": "2019-10-07T19:27:01+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6bec994a", "title": "“Adding Gradient Noise Improves Learning for Very Deep Networks”, Neelakantan et al.", "url": "https://davidstutz.de/adding-gradient-noise-improves-learning-for-very-deep-networks-neelakantan-et-al/", "published_at": "2019-10-05T19:19:10+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6c3719fa", "title": "“Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples”, Lee et al.", "url": "https://davidstutz.de/training-confidence-calibrated-classifiers-for-detecting-out-of-distribution-samples-lee-et-al/", "published_at": "2019-10-04T19:08:21+00:00" }, { "id": "01a0c582-b9df-7018-9b4d-237f6c845ea7", "title": "“The Limitations of Adversarial Training and the Blind-Spot Attack”, Zhang et al.", "url": "https://davidstutz.de/the-limitations-of-adversarial-training-and-the-blind-spot-attack-zhang-et-al/", "published_at": "2019-10-04T18:57:39+00:00" }, { "id": "01a0c590-d588-733e-a380-acb563c276ce", "title": "“A Theoretical Framework for Robustness of (Deep) Classifiers against Adversarial Samples”, Wang et al.", "url": "https://davidstutz.de/a-theoretical-framework-for-robustness-of-deep-classifiers-against-adversarial-samples-wang-et-al/", "published_at": "2019-10-03T18:50:26+00:00" }, { "id": "01a0c590-d588-733e-a380-acb564a7e6cd", "title": "“Towards Poisoning of Deep Learning Algorithms with Back-gradient Optimization”, Muñoz-González et al.", "url": "https://davidstutz.de/towards-poisoning-of-deep-learning-algorithms-with-back-gradient-optimization-munoz-gonzalez-et-al/", "published_at": "2019-10-02T18:40:15+00:00" }, { "id": "01a0c590-d588-733e-a380-acb564cf238e", "title": "“MagNet: A Two-Pronged Defense against Adversarial Examples”, Meng and Chen", "url": "https://davidstutz.de/magnet-a-two-pronged-defense-against-adversarial-examples-meng-and-chen/", "published_at": "2019-10-01T18:38:14+00:00" }, { "id": "01a0c590-d588-733e-a380-acb5659bf1f8", "title": "“UPSET and ANGRI : Breaking High Performance Image Classifiers”, Sarkar et al.", "url": "https://davidstutz.de/upset-and-angri-breaking-high-performance-image-classifiers-sarkar-et-al/", "published_at": "2019-09-25T19:49:04+00:00" }, { "id": "01a0c590-d588-733e-a380-acb565e97700", "title": "“On the importance of single directions for generalization”, Morcos et al.", "url": "https://davidstutz.de/on-the-importance-of-single-directions-for-generalization-morcos-et-al/", "published_at": "2019-09-23T19:46:33+00:00" }, { "id": "01a0c590-d588-733e-a380-acb566cf354c", "title": "“Improving Transferability of Adversarial Examples with Input Diversity”, Xie et al.", "url": "https://davidstutz.de/improving-transferability-of-adversarial-examples-with-input-diversity-xie-et-al/", "published_at": "2019-09-22T11:53:44+00:00" }, { "id": "01a0c590-d588-733e-a380-acb56724c424", "title": "“Improving Network Robustness against Adversarial Attacks with Compact Convolution”, Ranjan et al.", "url": "https://davidstutz.de/improving-network-robustness-against-adversarial-attacks-with-compact-convolution-ranjan-et-al/", "published_at": "2019-09-21T11:49:54+00:00" }, { "id": "01a0c590-d588-733e-a380-acb567b4f15c", "title": "“Regularizing Neural Networks by Penalizing Confident Output Distributions”, Pereyra", "url": "https://davidstutz.de/regularizing-neural-networks-by-penalizing-confident-output-distributions-pereyra/", "published_at": "2019-09-20T11:30:55+00:00" }, { "id": "01a0c590-d588-733e-a380-acb56809281f", "title": "“Adversarial Geometry and Lighting using a Differentiable Renderer”, Liu et al.", "url": "https://davidstutz.de/adversarial-geometry-and-lighting-using-a-differentiable-renderer-liu-et-al/", "published_at": "2019-09-19T11:23:37+00:00" }, { "id": "01a0c590-d588-733e-a380-acb568ec40b9", "title": "“Enhanced Attacks on Defensively Distilled Deep Neural Networks”, Liu et al.", "url": "https://davidstutz.de/enhanced-attacks-on-defensively-distilled-deep-neural-networks-liu-et-al/", "published_at": "2019-09-18T11:19:36+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e9794232aec", "title": "“Breaking Transferability of Adversarial Samples with Randomness”, Zhou et al.", "url": "https://davidstutz.de/7079-2/", "published_at": "2019-09-15T19:27:00+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e97944e87e7", "title": "“Cost-Sensitive Robustness against Adversarial Examples”, Zhang and Evans", "url": "https://davidstutz.de/cost-sensitive-robustness-against-adversarial-examples-zhang-and-evans/", "published_at": "2019-09-13T19:22:16+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e979487a4d1", "title": "“Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)”, Kim et al.", "url": "https://davidstutz.de/interpretability-beyond-feature-attribution-quantitative-testing-with-concept-activation-vectors-tcav-kim-et-al/", "published_at": "2019-09-11T19:05:56+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e9795190aa4", "title": "“Black-box Adversarial Attacks with Limited Queries and Information”, Ilyas", "url": "https://davidstutz.de/black-box-adversarial-attacks-with-limited-queries-and-information-ilyas/", "published_at": "2019-09-09T19:01:39+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e97951932d9", "title": "“On the Intriguing Connections of Regularization, Input Gradients and Transferability of Evasion and Poisoning Attacks”, Demontis et al.", "url": "https://davidstutz.de/on-the-intriguing-connections-of-regularization-input-gradients-and-transferability-of-evasion-and-poisoning-attacks-demontis-et-al/", "published_at": "2019-09-07T18:59:58+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e979555738c", "title": "“Attacks Meet Interpretability: Attribute-steered Detection of Adversarial Samples”, Tao et al.", "url": "https://davidstutz.de/attacks-meet-interpretability-attribute-steered-detection-of-adversarial-samples-tao-et-al/", "published_at": "2019-09-05T21:03:16+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e979612997f", "title": "“Adversarial Dropout for Supervised and Semi-Supervised Learning”, Park et al.", "url": "https://davidstutz.de/adversarial-dropout-for-supervised-and-semi-supervised-learning-park-et-al/", "published_at": "2019-09-04T20:51:43+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e9796d9d598", "title": "“Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks”, Liu et al.", "url": "https://davidstutz.de/fine-pruning-defending-against-backdooring-attacks-on-deep-neural-networks-liu-et-al/", "published_at": "2019-09-02T20:46:37+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e97970330da", "title": "“On the Geometry of Adversarial Examples”, Khoury and Hadfield-Menell", "url": "https://davidstutz.de/on-the-geometry-of-adversarial-examples-khoury-and-hadfield-menell/", "published_at": "2019-09-01T20:44:13+00:00" }, { "id": "01a0e796-993e-712f-bcb5-9e9797acca49", "title": "“The Limitations of Model Uncertainty in Adversarial Settings”, Grosse et al.", "url": "https://davidstutz.de/the-limitations-of-model-uncertainty-in-adversarial-settings-grosse-et-al/", "published_at": "2019-08-31T20:38:46+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6e7c56c8a", "title": "“Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples”, Dong et al.", "url": "https://davidstutz.de/towards-interpretable-deep-neural-networks-by-leveraging-adversarial-examples-dong-et-al/", "published_at": "2019-08-30T20:35:27+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6e8106435", "title": "“The Secret Sharer: Measuring Unintended Neural Network Memorization & Extracting Secrets”, Carlini et al.", "url": "https://davidstutz.de/the-secret-sharer-measuring-unintended-neural-network-memorization-extracting-secrets-carlini-et-al/", "published_at": "2019-08-28T20:28:56+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6e842aec6", "title": "“Mitigating Evasion Attacks to Deep Neural Networks via Region-based Classification”, Cao and Gong", "url": "https://davidstutz.de/mitigating-evasion-attacks-to-deep-neural-networks-via-region-based-classification-cao-and-gong/", "published_at": "2019-08-26T20:09:29+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6e8fb9f79", "title": "“Curriculum Adversarial Training”, Cai et al.", "url": "https://davidstutz.de/curriculum-adversarial-training-cai-et-al/", "published_at": "2019-08-23T20:06:32+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6e970919d", "title": "“AI2: Safety and Robustness Certification of Neural Networks with Abstract Interpretation”, Gehr et al.", "url": "https://davidstutz.de/ai2-safety-and-robustness-certification-of-neural-networks-with-abstract-interpretation-gehr-et-al/", "published_at": "2019-08-21T19:48:16+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6ea33d8e8", "title": "“Towards Robust Interpretability with Self-Explaining Neural Networks”, Alvarez-Melis and Jaakola", "url": "https://davidstutz.de/towards-robust-interpretability-with-self-explaining-neural-networks-alvarez-melis-and-jaakola/", "published_at": "2019-08-18T19:38:22+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6eaad3bbd", "title": "“Efficient Repair of Polluted Machine Learning Systems via Causal Unlearning”, Cao et al.", "url": "https://davidstutz.de/efficient-repair-of-polluted-machine-learning-systems-via-causal-unlearning-cao-et-al/", "published_at": "2019-08-16T19:49:30+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6eab76d31", "title": "“SoK: Science, Security and the Elusive Goal of Security as a Scientific Pursuit”, Herley and Oorschot", "url": "https://davidstutz.de/sok-science-security-and-the-elusive-goal-of-security-as-a-scientific-pursuit-herley-and-oorschot/", "published_at": "2019-08-15T19:45:34+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6eb9ec831", "title": "“Model-Reuse Attacks on Deep Learning Systems”, Ji et al.", "url": "https://davidstutz.de/model-reuse-attacks-on-deep-learning-systems-ji-et-al/", "published_at": "2019-08-12T19:24:08+00:00" }, { "id": "01a0e7b7-f8a3-7175-abb0-9bb6ec743f97", "title": "“Playing the Game of Universal Adversarial Perturbations”, Pérolat et al.", "url": "https://davidstutz.de/playing-the-game-of-universal-adversarial-perturbations-perolat-et-al/", "published_at": "2019-08-09T19:22:44+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276ba983069a", "title": "“Secure Kernel Machines against Evasion Attacks”, Russu et al.", "url": "https://davidstutz.de/secure-kernel-machines-against-evasion-attacks-russu-et-al/", "published_at": "2019-08-08T19:18:15+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276baa073dae", "title": "“Progressive Neural Networks”, Rusu et al.", "url": "https://davidstutz.de/progressive-neural-networks-rusu-et-al/", "published_at": "2019-08-05T19:15:08+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276baab946cf", "title": "“Are adversarial examples inevitable?”, Shafahi et al.", "url": "https://davidstutz.de/are-adversarial-examples-inevitable-shafahi-et-al/", "published_at": "2019-08-03T18:41:02+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276baae4ade3", "title": "“Label Smoothing and Logit Squeezing: A Replacement for Adversarial Training?”, Shafahi et al.", "url": "https://davidstutz.de/label-smoothing-and-logit-squeezing-a-replacement-for-adversarial-training-shafahi-et-al/", "published_at": "2019-08-01T18:37:38+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276baae4fd1b", "title": "“Universal Adversarial Training”, Shafahi et al.", "url": "https://davidstutz.de/universal-adversarial-training-shafahi-et-al/", "published_at": "2019-07-29T18:34:47+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276bab13fdd0", "title": "“On the Robustness of Convolutional Neural Networks to Internal Architecture and Weight Perturbations”, Cheney et al.", "url": "https://davidstutz.de/on-the-robustness-of-convolutional-neural-networks-to-internal-architecture-and-weight-perturbations-cheney-et-al/", "published_at": "2019-07-28T18:00:41+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276babcea227", "title": "“Adversarial Initialization – when your network performs the way I want”, Grosse et al.", "url": "https://davidstutz.de/adversarial-initialization-when-your-network-performs-the-way-i-want-grosse-et-al/", "published_at": "2019-07-27T17:54:20+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276babf3cc5c", "title": "“Fault injection attack on deep neural network”, Liu et al.", "url": "https://davidstutz.de/fault-injection-attack-on-deep-neural-network-liu-et-al/", "published_at": "2019-07-25T17:51:23+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276babfb7952", "title": "“Robustness of Generalized Learning Vector Quantization Models against Adversarial Attacks”, Saralajew et al.", "url": "https://davidstutz.de/robustness-of-generalized-learning-vector-quantization-models-against-adversarial-attacks-saralajew-et-al/", "published_at": "2019-07-23T17:31:32+00:00" }, { "id": "01a0e7df-627b-70b8-a695-276bac496911", "title": "“Protecting Intellectual Property of Deep Neural Networks with Watermarking”, Zhang et al.", "url": "https://davidstutz.de/protecting-intellectual-property-of-deep-neural-networks-with-watermarking-zhang-et-al/", "published_at": "2019-07-22T17:17:08+00:00" } ] posts Claim your blog
Back to davidstutz.de
Blog · corpus.blog/blogs/davidstutz.de/posts

davidstutz.de

davidstutz.de

2020

“Adversarial camera stickers: A physical camera-based attack on deep learning systems”, Li et al.original ↗
23 Mar 2020
“On Norm-Agnostic Robustness of Adversarial Training”, Li et al.original ↗
20 Mar 2020
“Adversarially Robust Distillation”, Goldblum et al.original ↗
19 Mar 2020
“Local Gradients Smoothing: Defense Against Localized Adversarial Attacks”, Naseer et al.original ↗
17 Mar 2020
Adversarial Examples Leave the Data Manifoldoriginal ↗
12 Mar 2020
Code Released: Confidence-Calibrated Adversarial Trainingoriginal ↗
1 Mar 2020
Talk on Confidence-Calibrated Adversarial Training at BCAI and Tübingen AI Centeroriginal ↗
28 Feb 2020
Updated ArXiv Pre-Print “Confidence-Calibrated Adversarial Training”original ↗
26 Feb 2020
On-Manifold Adversarial Examplesoriginal ↗
11 Feb 2020
FONTS: A Synthetic MNIST-Like Dataset with Known Manifoldoriginal ↗
8 Jan 2020

2019

240+ Papers on Adversarial Examples and Out-of-Distribution Detectionoriginal ↗
10 Dec 2019
A Short Introduction to Bayesian Neural Networksoriginal ↗
25 Nov 2019
AI and Deep Learning at the 7th Heidelberg Laureate Forum 2019original ↗
6 Nov 2019
“Robustness and generalization”, Xu and Mannororiginal ↗
2 Nov 2019
“Second-Order Adversarial Attack and Certifiable Robustness”, Li et al.original ↗
1 Nov 2019
“Certified Robustness to Adversarial Examples with Differential Privacy”, Lecuyer et al.original ↗
30 Oct 2019
More Examples for Working with Torchoriginal ↗
29 Oct 2019
“ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness”, Geirhos et al.original ↗
29 Oct 2019
“An Empirical Evaluation on Robustness and Uncertainty of Regularization Methods”, Chun et al.original ↗
28 Oct 2019
“Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet”, Brendel and Bethgeoriginal ↗
26 Oct 2019
“Towards Stable and Efficient Training of Verifiably Robust Neural Networks”, Zhang et al.original ↗
24 Oct 2019
“Efficient Neural Network Robustness Certification with General Activation Functions”, Zhang et al.original ↗
23 Oct 2019
“Generalization in Deep Networks: The Role of Distance from Initialization”, Nagarajan and Kolteroriginal ↗
22 Oct 2019
“Likelihood Ratios for Out-of-Distribution Detection”, Ren et al.original ↗
21 Oct 2019
“On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Models”, Gowal et al.original ↗
21 Oct 2019
“Batch Normalization is a Cause of Adversarial Vulnerability”, Galloway et al.original ↗
20 Oct 2019
“Detecting Extrapolation with Influence Functions”, Madras et al.original ↗
19 Oct 2019
“Radial basis function neural networks: a topical state-of-the-artsurvey”, Dash et al.original ↗
18 Oct 2019
“How Can We Be So Dense? The Benefits of Using Highly Sparse Representations”, Ahmad and Scheinkmanoriginal ↗
17 Oct 2019
“Deep-RBF Networks Revisited: Robust Classification with Rejection”, Zadeh et al.original ↗
16 Oct 2019
ArXiv Pre-Print “Confidence-Calibrated Adversarial Training”original ↗
16 Oct 2019
“Neural Networks with Structural Resistance to Adversarial Attacks”, De ALfarooriginal ↗
15 Oct 2019
“Adversarial Examples Are Not Bugs, They Are Features”, Ilyas et al.original ↗
13 Oct 2019
“Bit-Flip Attack: Crushing Neural Network withProgressive Bit Search”, Rakin et al.original ↗
12 Oct 2019
“The Lottery Ticket Hypothesis: Training Pruned Neural Networks”, Frankle and Carbinoriginal ↗
11 Oct 2019
“Certified Adversarial Robustness via Randomized Smoothing”, Cohen et al.original ↗
8 Oct 2019
“Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks”, Liang et al.original ↗
7 Oct 2019
“Adding Gradient Noise Improves Learning for Very Deep Networks”, Neelakantan et al.original ↗
5 Oct 2019
“Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples”, Lee et al.original ↗
4 Oct 2019
“The Limitations of Adversarial Training and the Blind-Spot Attack”, Zhang et al.original ↗
4 Oct 2019
“A Theoretical Framework for Robustness of (Deep) Classifiers against Adversarial Samples”, Wang et al.original ↗
3 Oct 2019
“Towards Poisoning of Deep Learning Algorithms with Back-gradient Optimization”, Muñoz-González et al.original ↗
2 Oct 2019
“MagNet: A Two-Pronged Defense against Adversarial Examples”, Meng and Chenoriginal ↗
1 Oct 2019
“UPSET and ANGRI : Breaking High Performance Image Classifiers”, Sarkar et al.original ↗
25 Sept 2019
“On the importance of single directions for generalization”, Morcos et al.original ↗
23 Sept 2019
“Improving Transferability of Adversarial Examples with Input Diversity”, Xie et al.original ↗
22 Sept 2019
“Improving Network Robustness against Adversarial Attacks with Compact Convolution”, Ranjan et al.original ↗
21 Sept 2019
“Regularizing Neural Networks by Penalizing Confident Output Distributions”, Pereyraoriginal ↗
20 Sept 2019
“Adversarial Geometry and Lighting using a Differentiable Renderer”, Liu et al.original ↗
19 Sept 2019
“Enhanced Attacks on Defensively Distilled Deep Neural Networks”, Liu et al.original ↗
18 Sept 2019
“Breaking Transferability of Adversarial Samples with Randomness”, Zhou et al.original ↗
15 Sept 2019
“Cost-Sensitive Robustness against Adversarial Examples”, Zhang and Evansoriginal ↗
13 Sept 2019
“Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)”, Kim et al.original ↗
11 Sept 2019
“Black-box Adversarial Attacks with Limited Queries and Information”, Ilyasoriginal ↗
9 Sept 2019
“On the Intriguing Connections of Regularization, Input Gradients and Transferability of Evasion and Poisoning Attacks”, Demontis et al.original ↗
7 Sept 2019
“Attacks Meet Interpretability: Attribute-steered Detection of Adversarial Samples”, Tao et al.original ↗
5 Sept 2019
“Adversarial Dropout for Supervised and Semi-Supervised Learning”, Park et al.original ↗
4 Sept 2019
“Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks”, Liu et al.original ↗
2 Sept 2019
“On the Geometry of Adversarial Examples”, Khoury and Hadfield-Menelloriginal ↗
1 Sept 2019
“The Limitations of Model Uncertainty in Adversarial Settings”, Grosse et al.original ↗
31 Aug 2019
“Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples”, Dong et al.original ↗
30 Aug 2019
“The Secret Sharer: Measuring Unintended Neural Network Memorization & Extracting Secrets”, Carlini et al.original ↗
28 Aug 2019
“Mitigating Evasion Attacks to Deep Neural Networks via Region-based Classification”, Cao and Gongoriginal ↗
26 Aug 2019
“Curriculum Adversarial Training”, Cai et al.original ↗
23 Aug 2019
“AI2: Safety and Robustness Certification of Neural Networks with Abstract Interpretation”, Gehr et al.original ↗
21 Aug 2019
“Towards Robust Interpretability with Self-Explaining Neural Networks”, Alvarez-Melis and Jaakolaoriginal ↗
18 Aug 2019
“Efficient Repair of Polluted Machine Learning Systems via Causal Unlearning”, Cao et al.original ↗
16 Aug 2019
“SoK: Science, Security and the Elusive Goal of Security as a Scientific Pursuit”, Herley and Oorschotoriginal ↗
15 Aug 2019
“Model-Reuse Attacks on Deep Learning Systems”, Ji et al.original ↗
12 Aug 2019
“Playing the Game of Universal Adversarial Perturbations”, Pérolat et al.original ↗
9 Aug 2019
“Secure Kernel Machines against Evasion Attacks”, Russu et al.original ↗
8 Aug 2019
“Progressive Neural Networks”, Rusu et al.original ↗
5 Aug 2019
“Are adversarial examples inevitable?”, Shafahi et al.original ↗
3 Aug 2019
“Label Smoothing and Logit Squeezing: A Replacement for Adversarial Training?”, Shafahi et al.original ↗
1 Aug 2019
“Universal Adversarial Training”, Shafahi et al.original ↗
29 Jul 2019
“On the Robustness of Convolutional Neural Networks to Internal Architecture and Weight Perturbations”, Cheney et al.original ↗
28 Jul 2019
“Adversarial Initialization – when your network performs the way I want”, Grosse et al.original ↗
27 Jul 2019
“Fault injection attack on deep neural network”, Liu et al.original ↗
25 Jul 2019
“Robustness of Generalized Learning Vector Quantization Models against Adversarial Attacks”, Saralajew et al.original ↗
23 Jul 2019
“Protecting Intellectual Property of Deep Neural Networks with Watermarking”, Zhang et al.original ↗
22 Jul 2019
Previous