Osbert Bastani

Artificial Intelligence · 2016–2024

A researcher in the Science corpus: 11 papers; trusts 194 researchers and is trusted by 38 researchers here, as inferred from the citation graph.

Subfields

• Artificial Intelligence — 220 citations into it; #6,272 of 35,345 by trust here, #5,970 by citations
• Computer Vision and Pattern Recognition — 91 citations into it
• Safety, Robustness & Interpretability — 78 citations into it; #565 of 2,251 by trust here, #514 by citations
• Reinforcement Learning — 77 citations into it; #1,250 of 2,242 by trust here, #1,728 by citations
• Control and Systems Engineering — 49 citations into it
• Statistics and Probability — 22 citations into it
• Computational Theory and Mathematics — 19 citations into it
• Evaluation & Benchmarks — 19 citations into it

Whom Osbert Bastani trusts overall

Their strongest outbound trust, strongest first, inferred from their citations.

1. Geoffrey E. Hinton
2. Leo Breiman
3. R. Caruana
4. Cynthia Rudin
5. Niko Speybroeck
6. Jimmy Ba
7. Richard S. Sutton
8. Alex Krizhevsky
9. Himabindu Lakkaraju
10. Yoshua Bengio

Who trusts Osbert Bastani overall

1. Swarat Chaudhuri
2. Amit Dhurandhar
3. Patrick Hall
4. Carolyn Kim
5. Emir Demirovi'c
6. Matthew Sotoudeh
7. Yacine Izza
8. Amina Adadi
9. Vaishak Belle
10. Piyawat Lertvittayakumjorn

Papers from Osbert Bastani's perspective

Ordered by how strongly the researchers Osbert Bastani trusts endorse each paper. Same corpus, a different order for every researcher.

1. ImageNet classification with deep convolutional neural networks
2. Deep Residual Learning for Image Recognition (Computer Vision and Pattern Recognition)
3. Adam: A Method for Stochastic Optimization (International Conference on Learning Representations)
4. Long Short-Term Memory
5. Attention is All you Need (Neural Information Processing Systems)
6. Gradient-based learning applied to document recognition
7. ImageNet: A large-scale hierarchical image database
8. ImageNet Large Scale Visual Recognition Challenge
9. Learning Multiple Layers of Features from Tiny Images
10. Very Deep Convolutional Networks for Large-Scale Image Recognition (International Conference on Learning Representations)

Their papers

• VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Training (2022) — 495 citations
• DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset (2024) — 905 citations
• PAC Confidence Predictions for Deep Neural Network Classifiers (2020) — 27 citations
• Eureka: Human-Level Reward Design via Coding Large Language Models (2023) — 673 citations
• "How do I fool you?": Manipulating User Trust via Misleading Black Box Explanations (2019) — 298 citations
• Learning Interpretable Models with Causal Guarantees (2019) — 19 citations
• Fairness with Dynamics (2019) — 24 citations
• Verifiable Reinforcement Learning via Policy Extraction (2018) — 407 citations
• Interpreting Blackbox Models via Model Extraction (2017) — 196 citations
• Interpretability via Model Extraction (2017) — 139 citations
• Measuring Neural Net Robustness with Constraints (2016) — 451 citations