- Rating
- 2
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 1
- Soundness
- 2
Read review
Paper summary
This paper studies multiplayer performative prediction games with incomplete information, where players have private information and their joint decisions affect the data distributions. The authors formulate the problem as a performative Bayesian game and introduce the notion of a performatively stable Bayesian equilibrium. They analyze the convergence of alternating decision optimization and belief updates under strong monotonicity and distribution sensitivity conditions. Based on these results, they propose an adaptive regularization algorithm and evaluate it in crude oil trading and ride-sharing market simulations.
Critical strengths
- The paper addresses an interesting extension of multiplayer performative prediction. Allowing players to have private information makes the setting more realistic and introduces additional challenges beyond the complete-information case.
- The combination of performative prediction and Bayesian games is well motivated. I particularly like the idea of studying decision updates and belief updates together, rather than treating the data distribution shift and information uncertainty separately.
- The theoretical analysis is a valuable part of the paper. The authors derive a sensitivity-based convergence condition and attempt to extend the analysis to finite-sample settings. This provides a useful starting point for understanding the interaction between performativity and incomplete information.
- The proposed regularization approach is relatively simple and intuitive. Adjusting the regularization strength according to distribution sensitivity is a reasonable way to stabilize the decision dynamics.
- The experiments cover two different market settings and several levels of distribution sensitivity. The results are encouraging, and the authors also provide detailed proofs and experimental settings in the appendix.
Critical weaknesses
- Bayesian equilibrium formulation. I found the equilibrium definition somewhat difficult to follow. In a Bayesian game, players generally choose strategies that depend on their private types, but Equations (2)–(5) mainly describe a fixed joint decision vector. The appendix later uses type-dependent decisions. It would help to clarify how these two representations are connected and what exactly the Bayesian equilibrium mapping operates on.
- Belief updates and information assumptions. Equation (7) uses a likelihood to update beliefs based on observed decisions, but I could not find a sufficiently clear explanation of how this likelihood is obtained when the distribution maps are unknown. The experimental examples also appear to rely on additional observable information to infer players' costs. Clarifying what each player knows and observes would make the incomplete-information setting more convincing.
- Convergence analysis. Theorem 3.7 is central to the paper, but some steps in the proof could be explained more carefully. In particular, the proof uses the proposed stable equilibrium as a fixed point before explicitly establishing its existence. It would help to clarify the assumptions that ensure the equilibrium mapping has a well-defined fixed point, especially regarding the decision space.
- Finite-sample guarantee. I have some concerns about Theorem 3.9. Assumption 3.8 is described as a finite-variance condition, but it actually imposes a uniform bound on gradient estimation error. The resulting bound also does not explicitly depend on the sample size. Moreover, it is not entirely clear how the proof handles the difference between the current and limiting beliefs. A more precise explanation of these points would strengthen the theoretical contribution.
- Connection between the algorithm and theory. Algorithm 1 appears to use individual best-response updates, while the convergence analysis assumes a Bayesian equilibrium is computed at each iteration. It is unclear whether these are equivalent in the proposed implementation. Also, the sensitivity estimate in line 10 is based on consecutive decisions, whereas Definition 3.1 requires a global bound. I would like to understand how the algorithm ensures that the theoretical condition is actually satisfied. The regularization and truthful-reporting guarantees in Theorem 4.1 and Appendix I could also be stated more carefully, since the former concerns an upper bound and the latter relies on informal incentive arguments.
- Experimental validation. The experiments are useful, but they rely on two specific market models with largely simulated distribution shifts. Most results focus on total revenue rather than directly measuring equilibrium convergence. Reporting belief convergence, equilibrium residuals, or estimated versus true sensitivity would provide stronger evidence for the theoretical claims. A clearer distinction between the use of real market data and simulated market dynamics would also help.
Questions for authors
- Could the authors clarify whether the proposed Bayesian equilibrium is defined over type-dependent strategies or fixed decision vectors? How is the likelihood in Equation (7) constructed, and what information is assumed to be observable to the players?
- Could the authors explain how Algorithm 1 satisfies the assumptions of Theorem 3.9 in practice? In particular, how is the Bayesian equilibrium computed at each iteration, and how can the local sensitivity estimate guarantee the required global sensitivity bound?
5 comments
Regarding the belief update and the information structure. We are grateful for the reviewer's observation and for the chance to clarify this point.
Belief update. The reviewer is correct that the distribution maps are unknown to players. is the conditional probability of observing the decision given the history and the type . Specifically, at time step , players make decisions based on the data samples collected from the data distributions induced by . Given this information, the players can update the belief about their types without requiring the distribution maps.
In the experiments, this likelihood is calculated with the realized cost:
-
Cournot competition on crude oil exportation. The type is the average cost, and the realized cost is .
- Player makes decisions based on , and the other players make decisions according to the expectation of cost .
- After observing the decision , the other players can compute the cost of player by , where (the parameters in demand function), and (the base export quantities) are public knowledge, and is the expectation of the other players' decisions under the public belief. Thus, only the cost is unknown and it is recovered from the observed decision.
- The likelihood of the observed cost sequence is then
-
Bertrand competition on the ride-share market. The type is the average cost vector, and the realized cost is .
- Player makes decisions based on , and the other players make decisions according to .
- After observing the decision , the other players can compute the cost of player by , where (the parameters in demand function), (the base price), and (the demand parameter) are public knowledge, and is the expectation of the competitor's decision under the public belief. Thus, only the cost is unknown and it is recovered from the observed decision.
- The likelihood of the observed cost sequence is then .
Information structure. For player , the observability is shown as follows:
- Observable information:
- its own type ,
- the decisions of all players ,
- the data samples collected by all players ,
- the history containing historical joint decisions and historical data samples of all players.
- Unobservable information:
- the other players' types ,
- the distribution maps of all players.
Revision.
- We have clarified that is the conditional probability of observing player 's decision given the type in Section 2.2 (lines 216-217).
Regarding the convergence analysis. We thank the reviewer for this helpful suggestion.
We now establish the existence of a performatively stable Bayesian equilibrium before the convergence analysis.
When , we prove the existence in the following steps:
- By Lemmas 3.6 and 3.7, we show that almost surely as .
- Since is compact, there exists a convergent subsequence .
- By and the triangle inequality, we have almost surely as .
- By and Lemmas 3.6 and 3.7, we have almost surely as .
- Since and , we have , which ensures the existence of a performatively stable Bayesian equilibrium.
Revision.
- We have added Assumption 3.4 (affine structure of the loss) in Section 3.2 (lines 283-295).
- We have proved the existence of a performatively stable Bayesian equilibrium in Appendix E (lines 868-939).
Regarding the finite-sample guarantee. We thank the reviewer for these insightful suggestions.
Bounded estimation error. We agree that the name "finite variance" was inaccurate. We have renamed Assumption 3.9 (original Assumption 3.8) to "Bounded estimation error", since it assumes a bound on the gradient estimation error without depending on sample size.
Sample-size dependence.
- Theorem 3.10 (original Theorem 3.9) is independent of the sample size , since the bound in Assumption 3.9 is a uniform bound on the gradient estimation error.
- Our additional Central Limit Theorem based analysis in Appendix G provides a convergence error bound that explicitly depends on the sample size . The average gradient estimation error with samples satisfies with probability under the large-sample assumption, where is the distribution with degrees of freedom and is the largest eigenvalue of the covariance matrix.
Current and limiting beliefs. We have revised the proof of Theorem 3.10 so that strong monotonicity is applied at the same belief. The gradient difference across beliefs vanishes almost surely as by Lemma 3.5 and the Portmanteau theorem.
Revision.
- We have renamed the assumption to "Bounded estimation error" (Assumption 3.9) in Section 3.3 (line 346).
- We have revised the proof to account for the gradients under different beliefs in Appendix F (lines 1066-1092).
We sincerely appreciate the reviewer's valuable time spent in carefully reviewing our manuscript, and the recognition of our motivation, theoretical analysis, algorithm design and experimental validation.
Below, we respond to the reviewer's valuable feedback. All revisions made in the manuscript are highlighted in blue for easy identification.
Regarding the Bayesian equilibrium formulation. We thank the reviewer for this insightful question.
We acknowledge that the simplification in the original manuscript may cause confusion. In the revised manuscript, we define the equilibrium explicitly in the strategy space:
Revision.