Research map / Learning theory and algorithms / Submodular and game algorithms
Bayesian Persuasion under Ex Ante and Ex Post Constraints
Yakov Babichenko, Inbal Talgam-Cohen, Konstantin Zabarnyi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Restrict each posterior to satisfy the ex post constraints | Bayesian persuasion · Constrained Bayesian persuasion, ex post | optimal scheme support size: at most k | Improved |
| Add m ex ante constraints | Bayesian persuasion · Constrained Bayesian persuasion, ex ante | optimal scheme support size: at most k+m; tight for every k and m | Mixed |
| Discretize posterior space and solve a finite LP | Ex ante-constrained signaling with constant k · Constrained Bayesian persuasion | additive utility error: at most ε | Improved |
| Allow constraint violations in the approximation | Bi-criteria FPTAS for ex ante-constrained signaling · Constrained Bayesian persuasion | ex ante constraint violation: at most ε | Mixed |
| Require a signaling scheme satisfying every constraint with strict inequality | Bi-criteria approximation · Constrained Bayesian persuasion | constraint satisfaction | Improved |
| Convert each of m convex ex ante constraints to its corresponding ex post constraint | Ex ante-constrained signaling · Constrained Bayesian persuasion | expected sender utility retained: at least 1/M^m | Hurt |
| Use corresponding ex post constraints instead of one convex ex ante constraint in the auction example | Welfare-maximizing single-item second-price auction · Online ad auction anti-discrimination example | welfare approximation: 1/2-approximation | Hurt |
| Use ex ante constraints rather than corresponding ex post constraints in the constructed instance | Constrained Bayesian persuasion · Two states, one convex constraint | multiplicative utility gap: gap tends to ∞ as ε tends to 0 | Improved |
Closest papers by meaning
- Bayesian Persuasion with Externalities: Exploiting Agent Types (AAAI 2025 · similarity 0.89)
- Online Bayesian Persuasion Without a Clue (NeurIPS 2024 · similarity 0.89)
- Computational Aspects of Bayesian Persuasion under Approximate Best Response (NeurIPS 2024 · similarity 0.87)
- Multi-Sender Persuasion: A Computational Perspective (ICML 2024 · similarity 0.86)
- Markov Persuasion Processes: Learning to Persuade From Scratch (NeurIPS 2025 · similarity 0.85)
- Learning to Persuade a Biased Receiver (NeurIPS 2026 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
