| dc.date.accessioned | 2024-07-17T12:04:34Z | |
| dc.date.available | 2024-07-17T12:04:34Z | |
| dc.date.issued | 2024 | |
| dc.identifier.uri | http://hdl.handle.net/10852/111490 | |
| dc.description.abstract | Machine learning is becoming vital in fields like healthcare, robotics, and online services. This dissertation studies how to design machine learning algorithms that effectively interact with and learn from a diverse group of agents (e.g., humans), which aim to assist, corrupt, or strategically game our algorithms. A central theme of this thesis is the alignment of an algorithm’s objective with that of its human user by actively seeking feedback for its actions. Moreover, we study how to design algorithms that learn and make robust decisions even when malicious agents attempt to impede the algorithm’s learning process. Finally, we study how machine learning algorithms can incentivize agents in online marketplaces to provide truthful information about their products. | en_US |
| dc.language.iso | en | en_US |
| dc.relation.haspart | Paper I. Interactive Inverse Reinforcement Learning for Cooperative Games. Thomas Kleine Buening, Anne-Marie George, Christos Dimitrakakis. In 39th International Conference on Machine Learning (ICML) 2022. The paper is included in the thesis. | |
| dc.relation.haspart | Paper II. Environment Design for Inverse Reinforcement Learning. Thomas Kleine Buening, Christos Dimitrakakis. Presented at the Human in the Loop Learning Workshop at NeurIPS 2022. The paper is included in the thesis. | |
| dc.relation.haspart | Paper III. ANACONDA: An Improved Dynamic Regret Algorithm for Adaptive Non-Stationary Dueling Bandits. Thomas Kleine Buening, Aadirupa Saha. In 26th International Conference on Artificial Intelligence and Statistics (AISTATS) 2023. The paper is included in the thesis. | |
| dc.relation.haspart | Paper IV. Minimax-Bayes Reinforcement Learning. Thomas Kleine Buening, Christos Dimitrakakis, Hannes Eriksson, Divya Grover, Emilio Jorge. In 26th International Conference on Artificial Intelligence and Statistics (AISTATS) 2023. The paper is included in the thesis. | |
| dc.relation.haspart | Paper V. Bandits Meet Mechanism Design to Combat Clickbait in Online Recommendation. Thomas Kleine Buening, Aadirupa Saha, Christos Dimitrakakis, Haifeng Xu. To appear in 12th International Conference on Learning Representations (ICLR) 2024. The paper is included in the thesis. | |
| dc.title | Learning in the Presence of Cooperative, Adversarial and Strategic Agents | en_US |
| dc.type | Doctoral thesis | en_US |
| dc.creator.author | Kleine Büning, Thomas | |
| dc.type.document | Doktoravhandling | en_US |