Bayesian persuasion in sequential decision-making
We study a dynamic model of Bayesian persuasion in sequential decision-making settings. An informed principal observes an external parameter of the world and advises an uninformed agent about actions to take over time. The agent takes actions in each time step based on the current state, the princip...
| Main Authors: | , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
AAAI Press
2022
|
| _version_ | 1797107262328668160 |
|---|---|
| author | Gan, J Majumdar, R Radanovic, G Singla, A |
| author_facet | Gan, J Majumdar, R Radanovic, G Singla, A |
| author_sort | Gan, J |
| collection | OXFORD |
| description | We study a dynamic model of Bayesian persuasion in sequential decision-making settings. An informed principal observes an external parameter of the world and advises an uninformed agent about actions to take over time. The agent takes actions in each time step based on the current state, the principal's advice/signal, and beliefs about the external parameter. The action of the agent updates the state according to a stochastic process. The model arises naturally in many applications, e.g., an app (the principal) can advice the user (the agent) on possible choices between actions based on additional real-time information the app has. We study the problem of designing a signaling strategy from the principal's point of view. We show that the principal has an optimal strategy against a myopic agent, who only optimizes their rewards locally, and the optimal strategy can be computed in polynomial time. In contrast, it is NP-hard to approximate an optimal policy against a far-sighted agent. Further, we show that if the principal has the power to threaten the agent by not providing future signals, then we can efficiently design a threat-based strategy. This strategy guarantees the principal's payoff as if playing against an agent who is far-sighted but myopic to future signals. |
| first_indexed | 2024-03-07T07:13:35Z |
| format | Conference item |
| id | oxford-uuid:3d94202a-3518-4ccf-9846-e6985e5351f8 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:13:35Z |
| publishDate | 2022 |
| publisher | AAAI Press |
| record_format | dspace |
| spelling | oxford-uuid:3d94202a-3518-4ccf-9846-e6985e5351f82022-07-22T09:57:27ZBayesian persuasion in sequential decision-makingConference itemhttp://purl.org/coar/resource_type/c_5794uuid:3d94202a-3518-4ccf-9846-e6985e5351f8EnglishSymplectic ElementsAAAI Press2022Gan, JMajumdar, RRadanovic, GSingla, AWe study a dynamic model of Bayesian persuasion in sequential decision-making settings. An informed principal observes an external parameter of the world and advises an uninformed agent about actions to take over time. The agent takes actions in each time step based on the current state, the principal's advice/signal, and beliefs about the external parameter. The action of the agent updates the state according to a stochastic process. The model arises naturally in many applications, e.g., an app (the principal) can advice the user (the agent) on possible choices between actions based on additional real-time information the app has. We study the problem of designing a signaling strategy from the principal's point of view. We show that the principal has an optimal strategy against a myopic agent, who only optimizes their rewards locally, and the optimal strategy can be computed in polynomial time. In contrast, it is NP-hard to approximate an optimal policy against a far-sighted agent. Further, we show that if the principal has the power to threaten the agent by not providing future signals, then we can efficiently design a threat-based strategy. This strategy guarantees the principal's payoff as if playing against an agent who is far-sighted but myopic to future signals. |
| spellingShingle | Gan, J Majumdar, R Radanovic, G Singla, A Bayesian persuasion in sequential decision-making |
| title | Bayesian persuasion in sequential decision-making |
| title_full | Bayesian persuasion in sequential decision-making |
| title_fullStr | Bayesian persuasion in sequential decision-making |
| title_full_unstemmed | Bayesian persuasion in sequential decision-making |
| title_short | Bayesian persuasion in sequential decision-making |
| title_sort | bayesian persuasion in sequential decision making |
| work_keys_str_mv | AT ganj bayesianpersuasioninsequentialdecisionmaking AT majumdarr bayesianpersuasioninsequentialdecisionmaking AT radanovicg bayesianpersuasioninsequentialdecisionmaking AT singlaa bayesianpersuasioninsequentialdecisionmaking |