More Like this

1. Online Learning with Optimism and Delay

   Flaspohler, Genevieve; Orabona, Francesco; Cohen, Judah; Mouatadid, Soukayna; Oprescu, Miruna; Orenstein, Paulo; Mackey, Lester (January 2021, Internation conference on machine learning)

   null (Ed.)

   Inspired by the demands of real-time climate and weather forecasting, we develop optimistic online learning algorithms that require no parameter tuning and have optimal regret guarantees under delayed feedback. Our algorithms -- DORM, DORM+, and AdaHedgeD -- arise from a novel reduction of delayed online learning to optimistic online learning that reveals how optimistic hints can mitigate the regret penalty caused by delay. We pair this delay-as-optimism perspective with a new analysis of optimistic learning that exposes its robustness to hinting errors and a new meta-algorithm for learning effective hinting strategies in the presence of delay. We conclude by benchmarking our algorithms on four subseasonal climate forecasting tasks, demonstrating low regret relative to state-of-the-art forecasting models. 

   more » « less
2. Learning Stress with Feet and Grids

   Lee, Seung Suk; Farinella, Alessa; Hughes, Cerys; Pater, Joe (January 2023, Proceedings of the 2022 Annual Meeting on Phonology)

   Elkins, Noah; Hayes, Bruce; Jo, Jinyoung; Siah, Jian-Leat (Ed.)
3. Online Learning with Optimism and Delay

   Flaspohler, Genevieve E; Orabona, Francesco; Cohen, Judah; Mouatadid, Soukayna; Oprescu, Miruna; Orenstein, Paulo; Mackey, Lester (July 2021, Proceedings of Machine Learning Research)

   Meila, Marina; Zhang, Tong (Ed.)

   Inspired by the demands of real-time climate and weather forecasting, we develop optimistic online learning algorithms that require no parameter tuning and have optimal regret guarantees under delayed feedback. Our algorithms—DORM, DORM+, and AdaHedgeD—arise from a novel reduction of delayed online learning to optimistic online learning that reveals how optimistic hints can mitigate the regret penalty caused by delay. We pair this delay-as-optimism perspective with a new analysis of optimistic learning that exposes its robustness to hinting errors and a new meta-algorithm for learning effective hinting strategies in the presence of delay. We conclude by benchmarking our algorithms on four subseasonal climate forecasting tasks, demonstrating low regret relative to state-of-the-art forecasting models. 

   more » « less
4. Learning When and Where to Zoom with Deep Reinforcement Learning

   https://doi.org/10.1109/CVPR42600.2020.01236  

   Uzkent, Burak; Ermon, Stefano (January 2020, IEEE Conference on Computer Vision and Pattern Recognition)

   null (Ed.)
5. Efficient Argument Structure Extraction with Transfer Learning and Active Learning

   https://doi.org/10.18653/v1/2022.findings-acl.36  

   Hua, X.; Wang, L. (January 2022, Findings of the Association for Computational Linguistics: ACL 2022)

   The automation of extracting argument structures faces a pair of challenges on (1) encoding long-term contexts to facilitate comprehensive understanding, and (2) improving data efficiency since constructing high-quality argument structures is time-consuming. In this work, we propose a novel context-aware Transformer-based argument structure prediction model which, on five different domains, significantly outperforms models that rely on features or only encode limited contexts. To tackle the difficulty of data annotation, we examine two complementary methods: (i) transfer learning to leverage existing annotated data to boost model performance in a new target domain, and (ii) active learning to strategically identify a small amount of samples for annotation. We further propose model-independent sample acquisition strategies, which can be generalized to diverse domains. With extensive experiments, we show that our simple-yet-effective acquisition strategies yield competitive results against three strong comparisons. Combined with transfer learning, substantial F1 score boost (5-25) can be further achieved during the early iterations of active learning across domains. 

   more » « less