Aleksander Madry

Machine Unlearning via Simulated Oracle Matching ICLR 2025 MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering ICLR 2025 Small-to-Large Generalization: Training Data Influences Models Consistently Across Scale ICLR 2025 ContextCite: Attributing Model Generation to Context NeurIPS 2024 Decomposing and Editing Predictions by Modeling Model Computation ICML 2024 DsDm: Model-Aware Dataset Selection with Datamodels ICML 2024 Improving Subgroup Robustness via Data Selection NeurIPS 2024 A Data-Based Perspective on Transfer Learning CVPR 2023 Distilling Model Failures as Directions in Latent Space ICLR 2023 FFCV: Accelerating Training by Removing Data Bottlenecks CVPR 2023 ModelDiff: A Framework for Comparing Learning Algorithms ICML 2023 Raising the Cost of Malicious AI-Powered Image Editing ICML 2023 Rethinking Backdoor Attacks ICML 2023 TRAK: Attributing Model Behavior at Scale ICML 2023 3DB: A Framework for Debugging Computer Vision Models NeurIPS 2022 Adversarially trained neural representations may already be as robust as corresponding biological neural representations ICML 2022 Certified Patch Robustness via Smoothed Vision Transformers CVPR 2022 Combining Diverse Feature Priors ICML 2022 Datamodels: Understanding Predictions with Data and Data with Predictions ICML 2022 Missingness Bias in Model Debugging ICLR 2022 BREEDS: Benchmarks for Subpopulation Shift ICLR 2021 Editing a classifier by rewriting its prediction rules NeurIPS 2021 Leveraging Sparse Linear Layers for Debuggable Deep Networks ICML 2021 Noise or Signal: The Role of Image Backgrounds in Object Recognition ICLR 2021 Unadversarial Examples: Designing Objects for Robust Vision NeurIPS 2021 A Closer Look at Deep Policy Gradients ICLR 2020 Do Adversarially Robust ImageNet Models Transfer Better? NeurIPS 2020 From ImageNet to Image Classification: Contextualizing Progress on Benchmarks ICML 2020 Identifying Statistical Bias in Dataset Replication ICML 2020 Implementation Matters in Deep RL: A Case Study on PPO and TRPO ICLR 2020 On Adaptive Attacks to Adversarial Example Defenses NeurIPS 2020 Adversarial Examples Are Not Bugs, They Are Features NeurIPS 2019 Exploring the Landscape of Spatial Robustness ICML 2019 Image Synthesis with a Single (Robust) Classifier NeurIPS 2019 Prior Convictions: Black-box Adversarial Attacks with Bandits and Priors ICLR 2019 Robustness May Be at Odds with Accuracy ICLR 2019 Training for Faster Adversarial Robustness Verification via Inducing ReLU Stability ICLR 2019 A Classification-Based Study of Covariate Shift in GAN Distributions ICML 2018 Adversarially Robust Generalization Requires More Data NeurIPS 2018 A Fast Algorithm for Separated Sparsity via Perturbed Lagrangians AISTATS 2018 How Does Batch Normalization Help Optimization? NeurIPS 2018 On the Limitations of First-Order Approximation in GAN Dynamics ICML 2018 Spectral Signatures in Backdoor Attacks NeurIPS 2018 Towards Deep Learning Models Resistant to Adversarial Attacks ICLR 2018