Summary
Many pivotal decisions hinge on navigating fundamental uncertainty, where a comprehensive description of the relevant contingencies and exact odds remains elusive. This project explores how such uncertainty influences learning, innovation, and strategic information sharing. While previous research has made significant strides in exploring information incentives, it predominantly revolves around scenarios with well-defined risks, neglecting the intricate real-world challenges of rapidly evolving, complex environments. My approach leverages decision-theoretic advances to construct novel theoretical frameworks of active learning and strategic disclosure, shedding light on the implications of fundamental uncertainty on the decision-makers' desire for robustness, demand for commitment, and strategic incentives.
The project entails the study of active experimentation, modeled as a canonical multi-arm bandit problem, in situations where the experimenter's knowledge about the underlying distribution of payoffs is limited. I propose a new theoretical approach centered on regret minimization and continual re-optimization. This work will generate new insights into the experimenter's dynamic tradeoff between exploitation and exploration in the face of shifting worst-case scenarios, as well as the feasibility of ex-ante optimal experimentation rules.
Shifting the focus to strategic interactions, I will investigate how uncertainty over the underlying process of information endowments impacts information-sharing incentives. My work will center on a canonical sender-receiver structure with verifiable information. I will examine the role of the receiver's sophistication in the predictions of such models, in particular, the inevitability of information unraveling. This will challenge some widely accepted insights of this literature.
The project entails the study of active experimentation, modeled as a canonical multi-arm bandit problem, in situations where the experimenter's knowledge about the underlying distribution of payoffs is limited. I propose a new theoretical approach centered on regret minimization and continual re-optimization. This work will generate new insights into the experimenter's dynamic tradeoff between exploitation and exploration in the face of shifting worst-case scenarios, as well as the feasibility of ex-ante optimal experimentation rules.
Shifting the focus to strategic interactions, I will investigate how uncertainty over the underlying process of information endowments impacts information-sharing incentives. My work will center on a canonical sender-receiver structure with verifiable information. I will examine the role of the receiver's sophistication in the predictions of such models, in particular, the inevitability of information unraveling. This will challenge some widely accepted insights of this literature.
Unfold all
/
Fold all
More information & hyperlinks
| Web resources: | https://cordis.europa.eu/project/id/101165999 |
| Start date: | 01-01-2025 |
| End date: | 31-12-2029 |
| Total budget - Public funding: | 1 402 678,00 Euro - 1 402 678,00 Euro |
Cordis data
Original description
Many pivotal decisions hinge on navigating fundamental uncertainty, where a comprehensive description of the relevant contingencies and exact odds remains elusive. This project explores how such uncertainty influences learning, innovation, and strategic information sharing. While previous research has made significant strides in exploring information incentives, it predominantly revolves around scenarios with well-defined risks, neglecting the intricate real-world challenges of rapidly evolving, complex environments. My approach leverages decision-theoretic advances to construct novel theoretical frameworks of active learning and strategic disclosure, shedding light on the implications of fundamental uncertainty on the decision-makers' desire for robustness, demand for commitment, and strategic incentives.The project entails the study of active experimentation, modeled as a canonical multi-arm bandit problem, in situations where the experimenter's knowledge about the underlying distribution of payoffs is limited. I propose a new theoretical approach centered on regret minimization and continual re-optimization. This work will generate new insights into the experimenter's dynamic tradeoff between exploitation and exploration in the face of shifting worst-case scenarios, as well as the feasibility of ex-ante optimal experimentation rules.
Shifting the focus to strategic interactions, I will investigate how uncertainty over the underlying process of information endowments impacts information-sharing incentives. My work will center on a canonical sender-receiver structure with verifiable information. I will examine the role of the receiver's sophistication in the predictions of such models, in particular, the inevitability of information unraveling. This will challenge some widely accepted insights of this literature.
Status
SIGNEDCall topic
ERC-2024-STGUpdate Date
20-11-2024
Images
No images available.
Geographical location(s)
