Profiles2020-2026
Ambuj Tewari
University of Michigan
16Papers
14Last author
15Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers16
- Continuum Transformers Perform In-Context Learning by Operator Gradient DescentAuthor 3 of 3ICLR 2026 · 3 citations
- Conformal Prediction for Ensembles: Improving Efficiency via Score-Based AggregationAuthor 3 of 3NeurIPS 2025 · 9 citations
- Generator-Mediated Bandits: Thompson Sampling for GenAI-Powered Adaptive InterventionsAuthor 4 of 4NeurIPS 2025 · 1 citation
- A Computationally Efficient Algorithm for Infinite-Horizon Average-Reward Linear MDPsAuthor 2 of 2ICML 2025
- On the Benefits of Active Data Collection in Operator LearningAuthor 2 of 2ICML 2025
- Online Classification with PredictionsAuthor 2 of 2NeurIPS 2024 · 9 citations
- A Primal-Dual Algorithm for Offline Constrained Reinforcement Learning with Linear MDPsAuthor 2 of 2ICML 2024 · 5 citations
- Smoothed Online Classification can be Harder than Batch ClassificationAuthor 3 of 3NeurIPS 2024 · 2 citations
- On the Computational Complexity of Private High-dimensional Model SelectionAuthor 3 of 3NeurIPS 2024
- Thompson Sampling for High-Dimensional Sparse Linear Contextual BanditsAuthor 3 of 3ICML 2023 · 15 citations
- Adaptive Sampling for DiscoveryAuthor 3 of 4NeurIPS 2022 · 5 citations
- Online Agnostic Multiclass BoostingAuthor 2 of 2NeurIPS 2022 · 3 citations
- Causal Bandits with Unknown Graph StructureAuthor 3 of 3NeurIPS 2021 · 53 citations
- Representation Learning Beyond Linear Prediction FunctionsAuthor 2 of 2NeurIPS 2021 · 27 citations
- TorsionNet: A Reinforcement Learning Approach to Sequential Conformer SearchAuthor 6 of 7NeurIPS 2020 · 33 citations
- On the Equivalence between Online and Private Learnability beyond Binary ClassificationAuthor 3 of 3NeurIPS 2020 · 18 citations
