| The effect of modeling human rationality level on learning rewards from multiple feedback types GR Ghosal, M Zurek, DS Brown, AD Dragan Proceedings of the AAAI Conference on Artificial Intelligence 37 (5), 5983-5992, 2023 | 36 | 2023 |
| A deep deterministic policy gradient based network scheduler for deadline-driven data transfers GR Ghosal, D Ghosal, A Sim, AV Thakur, K Wu 2020 IFIP Networking Conference (Networking), 253-261, 2020 | 17 | 2020 |
| Understanding Finetuning for Factual Knowledge Extraction GR Ghosal, T Hashimoto, A Raghunathan Proceedings of the 41st International Conference on Machine Learning 235 …, 2024 | 7 | 2024 |
| Multi-modal prototype learning for interpretable multivariable time series classification GR Ghosal, R Abbasi-Asl arXiv preprint arXiv:2106.09636, 2021 | 4 | 2021 |
| Failure Modes of LLMs for Causal Reasoning on Narratives K Yamin, S Gupta, GR Ghosal, ZC Lipton, B Wilder arXiv preprint arXiv:2410.23884, 2024 | 2 | 2024 |
| A generalized acquisition function for preference-based reward learning E Ellis, GR Ghosal, SJ Russell, A Dragan, E Bıyık 2024 IEEE International Conference on Robotics and Automation (ICRA), 2814-2821, 2024 | 2 | 2024 |
| Contextual Reliability: When Different Features Matter in Different Contexts GR Ghosal, A Setlur, DS Brown, A Dragan, A Raghunathan International Conference on Machine Learning 2023, 2023 | 2 | 2023 |
| Interpretable Prototype Discovery in Deep Learning-Based Time Series Classification GR Ghosal, R Abbasi-Asl Integrated Systems: Data Driven Engineering, 25-33, 2024 | | 2024 |
| Disentangling Sequence Memorization and General Capability in Large Language Models GR Ghosal, P Maini, A Raghunathan ICLR 2025 Workshop on Foundation Models in the Wild, 0 | | |
| Lowering Data Diversity can Accelerate Training: Case Studies in Synthetic Tasks S Kotha, U Girit, T Kumar, GR Ghosal, A Raghunathan | | |
Anca D DraganAssistant Professor at UC Berkeley // Director, AI Safety and Alignment, Google DeepMindVerified email at berkeley.edu
