arxiv: 2605.14193 · v1 · pith:OAAROTXHnew · submitted 2026-05-13 · 🧮 math.ST · stat.TH

Equilibrium and Pricing in Consumer Networks with Nonlinear Utilities: An Online Shape-Constrained Learning Approach

Daniele Bracale , George Michailidis This is my paper

Pith reviewed 2026-05-15 01:41 UTC · model grok-4.3

classification 🧮 math.ST stat.TH

keywords consumer networksmonopoly pricingnonlinear utilitiesnetwork equilibriumisotonic regressionno-regret learningnetwork influenceprice discrimination

0 comments

The pith

Consumer networks with nonlinear utilities admit a unique equilibrium that enables targeted monopoly pricing and online recovery of unknown demand functions.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that general nonlinear utility functions, such as logit, isoelastic, and Stone-Geary forms, produce a unique consumption equilibrium in directed and signed networks when contraction and variational conditions hold. This characterization lets a monopolist discriminate prices across communities and influencers by means of an extended influence measure that generalizes Katz-Bonacich centrality. When the utility functions themselves are unknown, the paper supplies a shape-constrained isotonic regression procedure that learns them from observed choices in an online fashion and proves strict no-regret convergence. The resulting framework therefore joins equilibrium theory directly to a nonparametric pricing algorithm that requires no tuning parameters.

Core claim

Under contraction and variational conditions on consumer utilities, a unique Nash equilibrium in consumption quantities exists even when network externalities are asymmetric and signed. This equilibrium supports analysis of optimal monopoly prices that incorporate a generalized network influence measure extending classical centrality concepts. When utilities are unknown, an online isotonic regression estimator recovers them with strict no-regret guarantees, allowing the monopolist to implement the derived pricing rule without prior parametric assumptions.

What carries the argument

A shape-constrained isotonic regression estimator that learns unknown nonlinear utilities online without tuning parameters while preserving the contraction properties needed for unique equilibrium.

If this is right

Targeted price discrimination becomes feasible in community-structured and influencer-driven markets once the generalized influence measure is computed from the equilibrium mapping.
The same equilibrium characterization applies directly to logit-type discrete choice, isoelastic, and Stone-Geary utilities under one set of conditions.
The isotonic learning procedure can be run continuously as new consumption data arrive, updating prices with no additional parameter tuning.
Signed and asymmetric peer effects are accommodated without requiring symmetry or non-negativity restrictions on the social graph.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same contraction-based argument may apply to other network goods such as platform adoption or content consumption where peer utilities are nonlinear.
Regulators could use the generalized influence measure to quantify market power concentration arising from network structure rather than from cost advantages alone.
Extensions to repeated pricing games would require verifying that the learned utilities remain contractive across periods.

Load-bearing premise

Consumer utility functions must satisfy contraction and variational conditions that guarantee a unique equilibrium.

What would settle it

A concrete network instance with nonlinear utilities violating the contraction condition that produces two or more distinct stable consumption equilibria, or an observed learning trajectory whose cumulative regret fails to converge to zero.

Figures

Figures reproduced from arXiv: 2605.14193 by Daniele Bracale, George Michailidis.

**Figure 1.** Figure 1: Average across the N sellers of the cumulative regret as a function of the horizon T ∈ T , displayed on a log–log scale. For each value of T, we run 10 independent repetitions and report 95% confidence intervals. The empirical slopes m’s are consistent with the theoretical regret rates predicted by our Theorem 4 (same or lower rate), which are respectively (following the legend from top to bottom): 0.75, N… view at source ↗

**Figure 2.** Figure 2: Average optimal prices across nodes in the influencer network. The influencer, indexed by [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗

**Figure 3.** Figure 3: Network associated with the matrix G. which network is represented in [PITH_FULL_IMAGE:figures/full_fig_p018_3.png] view at source ↗

**Figure 4.** Figure 4: Signed skew-symmetric triangle associated with the matrix [PITH_FULL_IMAGE:figures/full_fig_p018_4.png] view at source ↗

**Figure 5.** Figure 5: This plot show the regret convergence for [PITH_FULL_IMAGE:figures/full_fig_p051_5.png] view at source ↗

**Figure 6.** Figure 6: Isotonic regression estimate for varying [PITH_FULL_IMAGE:figures/full_fig_p052_6.png] view at source ↗

read the original abstract

We study optimal monopoly pricing over consumer networks governed by general nonlinear utilities. In our framework, a consumer's utility is jointly determined by an individualized price and the consumption choices of their peers, propagated through a directed and signed social graph. This formulation encapsulates a broad class of utility functions; it strictly generalizes the traditional linear-quadratic framework to include logit-type discrete choice, isoelastic, and Stone-Geary utilities under a single theoretical umbrella. We first establish the existence and uniqueness of the consumer-side equilibrium under general contraction and variational conditions, explicitly accommodating asymmetric and signed network externalities. Leveraging this equilibrium characterization, we analyze targeted price discrimination within community-structured and influencer-driven markets. To this end, we introduce a generalized measure of network influence that extends classical Katz-Bonacich centrality beyond the Euclidean domain. Finally, addressing the challenge of unknown consumer utility functions, we develop a shape-constrained, tuning-parameter-free learning approach utilizing isotonic regression, for which we establish strict no-regret convergence guarantees. Supported by extensive simulations, our results seamlessly integrate equilibrium analysis and nonparametric learning into a cohesive monopoly pricing framework.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper unifies nonlinear utility equilibria in signed networks with a tuning-free isotonic learner for monopoly pricing, but the contraction conditions are not explicitly verified beyond the linear-quadratic case.

read the letter

Colleague, the main thing here is a framework that handles consumer equilibrium on directed signed networks when utilities take logit, isoelastic, or Stone-Geary forms instead of just linear-quadratic. They prove existence and uniqueness under contraction and variational conditions, define a generalized influence measure that extends Katz-Bonacich, and then learn the unknown utility shape online via isotonic regression with no tuning parameters and strict no-regret guarantees. The simulations tie the pieces together for targeted pricing in community and influencer settings. What works is the clean integration of the equilibrium step with the learning step and the fact that the learning method avoids extra hyperparameters, which makes it more usable in practice. The generalized influence measure is a straightforward and useful extension. The soft spot is the reach of the existence theorem. The abstract states that the contraction and variational conditions cover the listed nonlinear families, yet only the linear-quadratic case receives concrete checking. For isoelastic utility with sigma greater than one, the best-response map can have Lipschitz constant exceeding one when the network is dense or externalities are negative, so uniqueness does not automatically carry over without explicit modulus bounds or safe parameter ranges for each family. The learning guarantees rest on more standard isotonic results and look less exposed. This is aimed at people in network economics, operations research, and platform design who want to move past linear-quadratic assumptions. A reader focused on applied pricing with social effects would get value from the equilibrium characterization and the learning approach once the contraction conditions are tightened. I would send it to peer review because the combination of theory and nonparametric learning is coherent enough to merit referee scrutiny even with the gap on the nonlinear cases.

Referee Report

1 major / 1 minor

Summary. The paper studies optimal monopoly pricing in consumer networks with general nonlinear utilities, including logit, isoelastic, and Stone-Geary types. It establishes existence and uniqueness of the consumer equilibrium under contraction and variational inequality conditions that allow for asymmetric and signed network externalities. The work analyzes targeted price discrimination using a generalized network influence measure extending Katz-Bonacich centrality, and proposes an online shape-constrained learning approach based on isotonic regression with strict no-regret guarantees, validated through simulations.

Significance. If the central claims hold, particularly the applicability of the contraction conditions to the nonlinear utility families, this paper makes a substantial contribution by generalizing the standard linear-quadratic network game framework to a broader class of utilities. The integration of equilibrium characterization with a tuning-parameter-free learning method for unknown utilities, along with the extended influence measure, provides a cohesive framework for pricing in social networks that could inform both theory and practice in mechanism design and data-driven pricing.

major comments (1)

[Theorem on equilibrium existence (likely §3 or Theorem 1)] The assertion that the contraction and variational conditions cover logit-type, isoelastic, and Stone-Geary utilities is not supported by explicit verification or Lipschitz constant bounds beyond the linear-quadratic case. For the isoelastic utility u(x) = x^{1-σ}/(1-σ) with σ>1, the best-response mapping may fail to be a contraction for dense networks or negative externalities, as the modulus can exceed 1. This is load-bearing for the uniqueness claim and requires either parameter restrictions or counterexamples to be addressed.

minor comments (1)

[Abstract] The term 'strict no-regret convergence guarantees' should be defined more precisely in the main text, including the specific regret bound and how it relates to the isotonic regression estimator.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the thorough review and constructive comments on our manuscript. The primary concern regarding explicit verification of the contraction conditions for the nonlinear utility families is addressed below. We will incorporate the suggested clarifications to strengthen the equilibrium uniqueness claims.

read point-by-point responses

Referee: The assertion that the contraction and variational conditions cover logit-type, isoelastic, and Stone-Geary utilities is not supported by explicit verification or Lipschitz constant bounds beyond the linear-quadratic case. For the isoelastic utility u(x) = x^{1-σ}/(1-σ) with σ>1, the best-response mapping may fail to be a contraction for dense networks or negative externalities, as the modulus can exceed 1. This is load-bearing for the uniqueness claim and requires either parameter restrictions or counterexamples to be addressed.

Authors: We agree that the original manuscript did not include explicit Lipschitz constant derivations or verification for the logit, isoelastic, and Stone-Geary families beyond the linear-quadratic case. Our general contraction and variational inequality framework is designed to apply when the best-response mapping satisfies the stated conditions, but we acknowledge the need for concrete bounds to support the claims for these specific utilities. In the revised version, we will add a new appendix subsection deriving the relevant Lipschitz moduli for each utility class, including sufficient conditions on parameters (e.g., bounds on σ for isoelastic utilities, network sparsity via spectral radius of the interaction matrix, and restrictions on signed externality magnitudes) that ensure the contraction modulus remains strictly less than 1. For the isoelastic case with σ > 1, we will explicitly discuss potential failure modes in dense networks or with strong negative externalities and provide either parameter restrictions guaranteeing uniqueness or illustrative counterexamples where the equilibrium may not be unique. These additions will be cross-referenced in the main text of Section 3 and supported by additional simulation checks. revision: yes

Circularity Check

0 steps flagged

No significant circularity; equilibrium and learning results rest on independent standard conditions

full rationale

The derivation establishes existence and uniqueness of equilibrium via contraction and variational inequality conditions that are standard mathematical tools (fixed-point theorems) and explicitly stated as general assumptions accommodating the listed nonlinear utility families. The nonparametric learning component relies on isotonic regression with no-regret convergence, which is a separate, established technique not derived from the equilibrium characterization. No equations reduce a claimed prediction to a fitted input by construction, no self-citation chains bear the central load, and no ansatz is smuggled via prior author work. The framework is self-contained against external benchmarks such as contraction mapping and shape-constrained regression, with no evident renaming of known results or self-definitional steps.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Only the abstract is available; specific free parameters, axioms, and invented entities cannot be extracted in detail. The central claims rest on unstated contraction/variational conditions for equilibrium and standard properties of isotonic regression.

axioms (1)

domain assumption Consumer utilities satisfy contraction and variational conditions guaranteeing unique equilibrium
Invoked in abstract as basis for existence and uniqueness results.

pith-pipeline@v0.9.0 · 5493 in / 1299 out tokens · 54259 ms · 2026-05-15T01:41:43.064906+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

136 extracted references · 136 canonical work pages · 2 internal anchors

[1]

Journal of econometrics , volume=

Semiparametric analysis of discrete response: Asymptotic properties of the maximum score estimator , author=. Journal of econometrics , volume=. 1985 , publisher=

work page 1985
[2]

Journal of econometrics , volume=

Maximum score estimation of the stochastic utility model of choice , author=. Journal of econometrics , volume=. 1975 , publisher=

work page 1975
[3]

Journal of labor economics , volume=

Are unemployment and out of the labor force behaviorally distinct labor force states? , author=. Journal of labor economics , volume=. 1983 , publisher=

work page 1983
[4]

International Conference on Artificial Intelligence and Statistics , pages=

Towards agnostic feature-based dynamic pricing: Linear policies vs linear valuation with unknown noise , author=. International Conference on Artificial Intelligence and Statistics , pages=. 2022 , organization=

work page 2022
[5]

44th Annual IEEE Symposium on Foundations of Computer Science, 2003

The value of knowing a demand curve: Bounds on regret for online posted-price auctions , author=. 44th Annual IEEE Symposium on Foundations of Computer Science, 2003. Proceedings. , pages=. 2003 , organization=

work page 2003
[6]

Operations research , volume=

Dynamic pricing without knowing the demand function: Risk bounds and near-optimal algorithms , author=. Operations research , volume=. 2009 , publisher=

work page 2009
[7]

Operations research , volume=

Dynamic pricing with an unknown demand model: Asymptotically optimal semi-myopic policies , author=. Operations research , volume=. 2014 , publisher=

work page 2014
[8]

Operations Research , volume=

Dynamic pricing and demand learning with limited price experimentation , author=. Operations Research , volume=. 2017 , publisher=

work page 2017
[9]

Algorithmic Learning Theory , pages=

Dynamic pricing with finitely many unknown valuations , author=. Algorithmic Learning Theory , pages=. 2019 , organization=

work page 2019
[10]

Management Science , volume=

Discontinuous demand functions: Estimation and pricing , author=. Management Science , volume=. 2020 , publisher=

work page 2020
[11]

Bernoulli , volume=

Least squares estimation in the monotone single index model , author=. Bernoulli , volume=. 2019 , publisher=

work page 2019
[12]

2012 , publisher=

Information bounds and nonparametric maximum likelihood estimation , author=. 2012 , publisher=

work page 2012
[13]

Monotone least squares and isotonic quantiles , author=

work page
[14]

Mathematics of Operations Research , volume=

Distribution-free contextual dynamic pricing , author=. Mathematics of Operations Research , volume=. 2024 , publisher=

work page 2024
[15]

Applied mathematics and Optimization , volume=

Projections onto order simplexes , author=. Applied mathematics and Optimization , volume=. 1984 , publisher=

work page 1984
[16]

Technometrics , volume=

Nearly-isotonic regression , author=. Technometrics , volume=. 2011 , publisher=

work page 2011
[17]

On extensions of the Brunn-Minkowski and Pr

Brascamp, Herm Jan and Lieb, Elliott H , journal=. On extensions of the Brunn-Minkowski and Pr. 1976 , publisher=

work page 1976
[18]

2006 , publisher=

Convex functions and their applications , author=. 2006 , publisher=

work page 2006
[19]

International Workshop on Approximation Algorithms for Combinatorial Optimization , pages=

Sampling s-concave functions: The limit of convexity based isoperimetry , author=. International Workshop on Approximation Algorithms for Combinatorial Optimization , pages=. 2009 , organization=

work page 2009
[20]

Li, Shukai and Shi, Cong and Mehrotra, Sanjay , journal=

work page
[21]

Production and Operations Management , volume=

Competitive Demand Learning: A Noncooperative Pricing Algorithm with Coordinated Price Experimentation , author=. Production and Operations Management , volume=. 2024 , publisher=

work page 2024
[22]

Management science , volume=

Optimal dynamic pricing of inventories with stochastic demand over finite horizons , author=. Management science , volume=. 1994 , publisher=

work page 1994
[23]

European Journal of Operational Research , volume=

Optimal dynamic pricing of inventories with stochastic demand and discounted criterion , author=. European Journal of Operational Research , volume=. 2012 , publisher=

work page 2012
[24]

Gaussian

A generalized linear model with “Gaussian” regressor variables , author=. Selected Works of David Brillinger , pages=. 2012 , publisher=

work page 2012
[25]

International Statistical Review/Revue Internationale de Statistique , pages=

Elliptically symmetric distributions: A review and bibliography , author=. International Statistical Review/Revue Internationale de Statistique , pages=. 1981 , publisher=

work page 1981
[26]

Adaptive economic models , pages=

Learning by firms about demand conditions , author=. Adaptive economic models , pages=. 1975 , publisher=

work page 1975
[27]

Operations research , volume=

Learning and pricing with models that do not explicitly incorporate competition , author=. Operations research , volume=. 2015 , publisher=

work page 2015
[28]

Available at SSRN 4572453 , year=

Learning to price under competition for multinomial logit demand , author=. Available at SSRN 4572453 , year=

work page
[29]

Available at SSRN 4722908 , year=

Adaptive Learning in Uncertain and Sequential Competition , author=. Available at SSRN 4722908 , year=

work page
[30]

Energy Economics , volume=

Nonlinear inverse demand curves in electricity market modeling , author=. Energy Economics , volume=. 2022 , publisher=

work page 2022
[31]

Advances in Applied Energy , volume=

Conditions for profitable operation of P2X energy hubs to meet local demand with energy market access , author=. Advances in Applied Energy , volume=. 2023 , publisher=

work page 2023
[32]

Manufacturing & Service Operations Management , volume=

Price competition with the attraction demand model: Existence of unique equilibrium and its stability , author=. Manufacturing & Service Operations Management , volume=. 2006 , publisher=

work page 2006
[33]

IEEE Transactions on Information Theory , volume=

Structured signal recovery from non-linear and heavy-tailed measurements , author=. IEEE Transactions on Information Theory , volume=. 2018 , publisher=

work page 2018
[34]

Foundations of Computational mathematics , volume=

The convex geometry of linear inverse problems , author=. Foundations of Computational mathematics , volume=. 2012 , publisher=

work page 2012
[35]

Operations Research , year=

To interfere or not to interfere: Information revelation and price-setting incentives in a multiagent learning environment , author=. Operations Research , year=

work page
[36]

Network science, nonlinear science and infrastructure systems , pages=

Modeling the transient nature of dynamic pricing with demand learning in a competitive environment , author=. Network science, nonlinear science and infrastructure systems , pages=. 2007 , publisher=

work page 2007
[37]

Recherches sur les principes math

Cournot, AA , year=. Recherches sur les principes math

work page
[38]

Theorie mathematique de la richesse sociale

Review of “Theorie mathematique de la richesse sociale” and of “Recherches sur les principles mathematiques de la theorie des richesses.” , author=. Journal de savants , volume=

work page
[39]

Management Science , volume=

Price competition under mixed multinomial logit demand functions , author=. Management Science , volume=. 2013 , publisher=

work page 2013
[40]

Operations Research , volume=

Multiproduct price optimization and competition under the nested logit model with product-differentiated price sensitivities , author=. Operations Research , volume=. 2014 , publisher=

work page 2014
[41]

Operations research , volume=

Fixed-point approaches to computing Bertrand-Nash equilibrium prices under mixed-logit demand , author=. Operations research , volume=. 2011 , publisher=

work page 2011
[42]

Management Science , volume=

Dynamic pricing of perishable assets under competition , author=. Management Science , volume=. 2014 , publisher=

work page 2014
[43]

Operations Research , volume=

Multi-product price and assortment competition , author=. Operations Research , volume=. 2015 , publisher=

work page 2015
[44]

Operations Research , volume=

Duopoly competition with network effects in discrete choice models , author=. Operations Research , volume=. 2021 , publisher=

work page 2021
[45]

Journal of the American Statistical Association , volume=

Policy optimization using semiparametric models for dynamic pricing , author=. Journal of the American Statistical Association , volume=. 2024 , publisher=

work page 2024
[46]

Journal of Machine Learning Research , volume=

Perishability of data: dynamic pricing under varying-coefficient models , author=. Journal of Machine Learning Research , volume=

work page
[47]

Parametric and nonparametric models and methods in financial econometrics , author=

work page
[48]

Annals of statistics , volume=

Nonparametric estimation of multivariate convex-transformed densities , author=. Annals of statistics , volume=. 2010 , publisher=

work page 2010
[49]

Conference on Learning Theory , pages=

Adaptive learning in continuous games: Optimal regret bounds and convergence to nash equilibrium , author=. Conference on Learning Theory , pages=. 2021 , organization=

work page 2021
[50]

Operations research , volume=

Non-stationary stochastic optimization , author=. Operations research , volume=. 2015 , publisher=

work page 2015
[51]

Operations Research , volume=

Nonstationary stochastic optimization under L p, q-variation measures , author=. Operations Research , volume=. 2019 , publisher=

work page 2019
[52]

Operations Research , year=

Adaptive, doubly optimal no-regret learning in strongly monotone and exp-concave games with gradient feedback , author=. Operations Research , year=

work page
[53]

Management Science , volume=

Mostly exploration-free algorithms for contextual bandits , author=. Management Science , volume=. 2021 , publisher=

work page 2021
[54]

2024 , publisher=

Elliptically symmetric distributions in signal processing and machine learning , author=. 2024 , publisher=

work page 2024
[55]

Journal of the American Statistical Association , volume=

Sliced inverse regression for dimension reduction , author=. Journal of the American Statistical Association , volume=. 1991 , publisher=

work page 1991
[56]

Advances in Neural Information Processing Systems , volume=

Weighted linear bandits for non-stationary environments , author=. Advances in Neural Information Processing Systems , volume=

work page
[57]

arXiv preprint arXiv:2010.10182 , year=

The elliptical potential lemma revisited , author=. arXiv preprint arXiv:2010.10182 , year=

work page arXiv 2010
[58]

Proceedings of Machine Learning Research vol , volume=

Optimal regret bounds for generalized linear bandits under parameter drift , author=. Proceedings of Machine Learning Research vol , volume=

work page
[59]

International Conference on Machine Learning , pages=

Provably optimal algorithms for generalized linear contextual bandits , author=. International Conference on Machine Learning , pages=. 2017 , organization=

work page 2017
[60]

Foundations of computational mathematics , volume=

User-friendly tail bounds for sums of random matrices , author=. Foundations of computational mathematics , volume=. 2012 , publisher=

work page 2012
[61]

arXiv preprint arXiv:2003.10113 , year=

Algorithms for non-stationary generalized linear bandits , author=. arXiv preprint arXiv:2003.10113 , year=

work page arXiv 2003
[62]

Proceedings of the fourteenth international conference on artificial intelligence and statistics , pages=

Contextual bandits with linear payoff functions , author=. Proceedings of the fourteenth international conference on artificial intelligence and statistics , pages=. 2011 , organization=

work page 2011
[63]

Advances in neural information processing systems , volume=

Improved algorithms for linear stochastic bandits , author=. Advances in neural information processing systems , volume=

work page
[64]

International conference on machine learning , pages=

Thompson sampling for contextual bandits with linear payoffs , author=. International conference on machine learning , pages=. 2013 , organization=

work page 2013
[65]

International Conference on Machine Learning , pages=

Thompson sampling for high-dimensional sparse linear contextual bandits , author=. International Conference on Machine Learning , pages=. 2023 , organization=

work page 2023
[66]

Proceedings of the AAAI Conference on Artificial Intelligence , volume=

Double doubly robust thompson sampling for generalized linear contextual bandits , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

work page
[67]

Journal of Economic Theory , volume=

Cournot duopoly with two production periods , author=. Journal of Economic Theory , volume=. 1987 , publisher=

work page 1987
[68]

arXiv preprint arXiv:2303.03348 , year=

Thompson Sampling for Linear Bandit Problems with Normal-Gamma Priors , author=. arXiv preprint arXiv:2303.03348 , year=

work page arXiv
[69]

Artificial Intelligence and Statistics , pages=

Linear thompson sampling revisited , author=. Artificial Intelligence and Statistics , pages=. 2017 , organization=

work page 2017
[70]

IEEE Transactions on Information Theory , volume=

Real and complex monotone communication games , author=. IEEE Transactions on Information Theory , volume=. 2014 , publisher=

work page 2014
[71]

2024 , publisher=

Variational Inequalities in Management Science and Finance , author=. 2024 , publisher=

work page 2024
[72]

Revenue Maximization Under Sequential Price Competition Via The Estimation Of s-Concave Demand Functions

Revenue Maximization Under Sequential Price Competition Via The Estimation Of s-Concave Demand Functions , author=. arXiv preprint arXiv:2503.16737 , year=

work page internal anchor Pith review Pith/arXiv arXiv
[73]

Journal of machine learning research , volume=

Nash Q-learning for general-sum stochastic games , author=. Journal of machine learning research , volume=

work page
[74]

Handbook of reinforcement learning and control , pages=

Multi-agent reinforcement learning: A selective overview of theories and algorithms , author=. Handbook of reinforcement learning and control , pages=. 2021 , publisher=

work page 2021
[75]

arXiv preprint arXiv:2011.00583 , year=

An overview of multi-agent reinforcement learning from game theoretical perspective , author=. arXiv preprint arXiv:2011.00583 , year=

work page arXiv 2011
[76]

Advances in neural information processing systems , volume=

Parametric bandits: The generalized linear case , author=. Advances in neural information processing systems , volume=

work page
[77]

21st Annual Conference on Learning Theory , number=

Stochastic linear optimization under bandit feedback , author=. 21st Annual Conference on Learning Theory , number=

work page
[78]

Journal of the Operational Research Society , pages=

Dynamic price competition and promotion strategy with cross-network effect: Implications for e-commerce platforms , author=. Journal of the Operational Research Society , pages=. 2025 , publisher=

work page 2025
[79]

Mathematical and computational models for congestion charging , pages=

Dynamic pricing: A learning approach , author=. Mathematical and computational models for congestion charging , pages=. 2006 , publisher=

work page 2006
[80]

Proceedings of the American mathematical society , volume=

Statistical limit points , author=. Proceedings of the American mathematical society , volume=

work page

Showing first 80 references.