betaFeedback

WorksMathematics

Optimal Learning in Fairness-Aware Contextual Bandits

Abstract

Mitigating unfair disparities in sequential decision-making is a growing concern in machine learning. We study contextual bandits under a global covariate-based Lipschitz fairness constraint, pioneered by \cite{dwork2012fairness}, requiring individuals with similar contexts to receive similar action distributions. We show that this constraint induces a randomisation region near the decision boundary and explicitly quantify the population level cost of fairness. We further establish a canonical representation of the optimal fair policy, which reduces the fair policy design to a lower-dimensional boundary optimisation problem. We develop a lazy fair greedy algorithm that exploits fairness-induced covariate diversity under continuous contexts, with policy updates computed via finite-dimensional linear programming. For general context distributions, including discrete and mixed contexts, we propose a lazy polyhedral upper confidence bound (UCB) algorithm, whose optimistic optimisation reduces to finitely many linear programs over confidence-set vertices. Both methods satisfy fairness at every round and achieve regret of order O(\sqrt{T} + T^{(d-2)/(d-1)}{\indc\{d\geq 4\}}) up to poly-logarithmic factors, where dd is the dimension of the contexts. A matching minimax lower bound for ℓ1\ell_1-based fairness establishes optimality, and reveals an unavoidable nonparametric cost of learning the decision boundary, which dominates the parametric learning cost when d≥4d \geq 4. Experiments on synthetic and real data support our theoretical findings and demonstrate the practical effectiveness of our methods.

Introduction

Multi-armed bandits (MABs) provide a fundamental framework for interactive decision-making, in which a policymaker repeatedly selects actions, observes feedback, and adapts its strategy based on past observations. In many modern applications, the policymaker also has access to covariate information that can inform its decisions beyond the feedback collected from previous actions. This naturally motivates the study of contextual MABs, which have become a central framework for adaptive sequential policy designs, with successful applications in healthcare and clinical trials ([3]), online recommendation ([59]) and finance ([9]), among others.

In many applications, designing policies solely to maximise rewards, however, may raise serious ethical or even legal concerns. Empirical evidence from online advertising demonstrates that optimisation-driven delivery systems can produce unequal exposure across demographic groups, even in the absence of explicitly discriminatory targeting ([2]). These concerns are also recognised in broader AI governance frameworks. For instance, the National Institute of Standards and Technology’s AI Risk Management Framework of the U.S. emphasises the importance of identifying and mitigating bias and potential harms to individuals, organisations and society through the design:VEVENT and deployment of AI systems (National Institute of Standards and Technology, [41]). Similarly, China’s Provisions on the Administration of Algorithmic Recommendation for Internet Information Services explicitly prohibit the use of algorithmic recommendations to impose unreasonable differential pricing based on consumers’ preferences and transaction histories (Cyberspace Administration of China et al., [19]). In response, algorithmic fairness has emerged as a critical consideration in both research and practice, motivating substantial efforts to mitigate bias in existing machine learning methods and to embed explicit fairness guarantees into sequential decision-making policies.

Fairness in decision-making has been studied from various perspectives, with two prominent approaches focusing on group fairness and individual fairness. Group fairness concerns disparities across groups defined by sensitive attributes, such as race or gender, and typically requires certain statistical measures, such as selection rates or false positive rates, to be comparable across groups (e.g. Zeng et al., [58]; Hou and Zhang, [29]; Fukuchi and Sakuma, [23]). Individual fairness instead requires individuals who are similar with respect to the task at hand to receive similar treatment, a property not guaranteed by group-level criteria (Dwork et al., [21]).

Individual fairness is particularly relevant to contextual bandit problems, where decisions are made sequentially based on individuals’ covariates. In automated lending, for example, an unconstrained reward-maximising policy may deterministically approve one applicant and reject another with a nearly identical financial profile simply because they lie on opposite sides of a decision boundary. This discontinuity underscores the importance of treating similar individuals similarly, as advocated by the fairness through awareness principle of Dwork et al. [21].

In this paper, we study contextual bandits under a global covariate-based Lipschitz fairness constraint, requiring individuals with similar contexts to receive similar action selection probabilities. Imposing this constraint fundamentally changes both the oracle problem structure and the statistical complexity of bandit learning. Without fairness constraints, the optimal policy in a linear contextual bandit is determined pointwise by the sign of a linear reward gap between the two arms, yielding a parametric learning problem whose minimax regret scales as T\sqrt{T} up to logarithmic factors (Lattimore and Szepesvári, [33]). Under Lipschitz fairness, by contrast, a policy cannot switch abruptly across the decision boundary. Instead, it must transition smoothly across a neighbourhood of the boundary, and the optimal shape of this transition depends on how covariate probability mass is distributed near the boundary. Consequently, even under a linear reward model, learning an optimal fair policy entails a nonparametric functional optimisation problem along the (d−1)(d - 1)-dimensional decision boundary. We formalise this phenomenon by characterising the geometry and canonical structure of optimal fair policies, quantifying the population-level cost of fairness, and establishing matching upper and minimax lower regret bounds.

Related literature

To highlight the contribution of this paper, we briefly summarise the literature most closely related to fair bandits and interactive decision making in this subsection.

A prominent strand of the fair bandit literature imposes arm-level allocation constraints, requiring each arm to receive a minimum share of selections. Chen et al. [16] study contextual MABs with constraints that guarantee the marginal probability of each arm being pulled is uniformly lower-bounded. Patil et al. [42] consider stochastic Bernoulli bandits with per-arm quota constraints, requiring each arm to be pulled at least a prescribed fraction of rounds. In the combinatorial bandit setting, where users can choose a combination of arms per round, Li et al. [34] enforce asymptotic minimum pull-proportion constraints using virtual queues, while Xu et al. (2020) impose lower bounds on the empirical selection proportion of each arm throughout the time horizon. Furthermore, Liu et al. [36] develop a more general framework for combinatorial bandits with linear constraints, which includes Xu et al. (2020) as a special case. Wu and Li [55] consider Bernoulli bandits with minimum average reward requirements for each arm, shifting the requirement from selection frequency. to realised reward. Crucially, these formulations enforce fairness across arms by guaranteeing minimum exposure or reward per action, rather than across individuals with varying covariates.

Another related line of work studies individual fairness through meritocratic or similarity-based constraints on the policy. Joseph et al. [32] introduce a meritocratic notion of fairness, requiring that an arm should not be selected with higher probability than another arm unless it has higher expected reward. This meritocratic perspective is further developed by Joseph et al. [31], studying fairness in contextual bandits with infinite convex action sets. In a similar spirit, Wang et al. [52] study fairness of exposure, under which the probability of selecting an arm is required to be proportional to an application-dependent merit function of its expected reward. Moving from merit-based exposure to similarity-based constraints, Liu et al. [37] study smooth and calibrated fairness in non-contextual MABs, where selection probabilities are constrained by similarities or divergences between reward distributions. Gillen et al. [24] consider linear contextual bandits with Lipschitz-type individual fairness constraints. Their constraints compare arm-specific contexts within each single round rather than coupling treatment probabilities globally across individuals arriving over the covariate space.

Other approaches incorporate fairness into the learning objective, rather than imposing it as an explicit constraint. For example, Blanchard and Goyal [7] study collaborative bandits using min-max regret as a fairness-aware objective. Similar objective based formulations also include Nash regret in Barman et al. [4] and Sawarni et al. [46], and α\alpha-fair utility functions in Chaudhary and Sinha [11]. Group fairness has also been studied in contextual bandits. Grazzi et al. [26] consider linear contextual bandits with a group meritocratic fairness constraint based on sensitive-group rankings. This notion is closely related to the meritocratic fairness concepts of Joseph et al. [32] and Wang et al. [52], but applies fairness at the level of sensitive groups rather than individual arms. These group-level criteria do not generally ensure similar action probabilities for individuals with similar contexts, as required by our individual fairness constraint.

Beyond bandits, individual fairness has also been studied in other interactive decision-making problems. Gupta and Kamble [27] introduce a notion named individual fairness in hindsight, comparing current and past decisions through a one-sided similarity constraint, and establish sublinear regret. Salem et al. [45] study envy-freeness across groups and time. In contextual dynamic pricing, Chen et al. [15] require similar prices for customers with similar baseline utilities, and characterise optimal fair pricing policies and the regret of demand learning. Although related in spirit, their constraint acts directly on continuous prices, whereas ours acts on selection probabilities over discrete actions, inducing fundamentally different boundary geometry and policy-learning challenges that preclude a direct application of their techniques. Other pricing studies focus on disparities across customer groups and over time. Cohen et al. [18] consider price and demand fairness, while Chen et al. [13] constrain relative price differences between groups under nonparametric demand. In online resource allocation, Chen et al. [12] formulate the problem as a linear program and use the analytic centre of its offline optimal solution set to guide fair online allocations.

In this paper, we focus on individual fairness across contexts, formalised through a global Lipschitz constraint on action selection probabilities in Definition 1, with more discussions in Section 2.1.

List of contributions

We summarise the main contributions of the paper.

  • To the best of our knowledge, this is the first paper to systematically study contextual bandits under a global covariate-based Lipschitz fairness constraint (Definition 1), extending the individual fairness concepts in Dwork et al. [21] to bandit learning. Our framework incorporates similarity-based fairness into contextual bandit problems, requiring the smoothness of the policy with respect to the underlying context space.

  • At the population level, we demonstrate that the Lipschitz fairness fundamentally reshapes the oracle policy by inducing a randomisation region around the decision boundary. The induced ran-

domisation leads to a deviation from the unconstrained reward-maximising policy. We use this deviation to explicitly quantify the population-level cost of fairness, i.e. the expected reward loss relative to the unconstrained oracle (Theorem 2).

  • We establish a canonical representation of optimal fair policies (Theorem 3), showing how a policy’s value on the decision boundary determines its reward-maximising Lipschitz extension over the entire context space. Moreover, we show that the resulting infinite-dimensional empirical boundary optimisation reduces without relaxation gap to a finite-dimensional linear program. Guided by this structure, we develop a lazy adaptive greedy algorithm for continuous contexts that exploits the fairness induced exploration (Algorithms 1 and 5). For more general context distributions, we develop a lazy polyhedral upper confidence bound (UCB) algorithm with efficient vertex-based optimisation (Algorithms 3 and 6).

  • We show that both proposed algorithms satisfy the Lipschitz fairness constraint without slackness and attain a cumulative regret bound of order T+T(d−2)/(d−1)1(d≥4)\sqrt{T}+T^{(d-2)/(d-1)}\mathbf{1}(d\ge4), up to poly-logarithmic factors (Theorem 4 and 6). The additional dimension-dependent term reflects the complexity of learning over the class of Lipschitz functions defined on the (d−1)(d-1)-dimensional decision boundary, arising from the Lipschitz fairness constraints. We further establish a matching minimax lower bound (Theorem 7), demonstrating the rate optimality of both proposed algorithms and that, whenever d≥4d\ge4, Lipschitz fairness incurs an unavoidable nonparametric dimension-dependent cost that dominates the intrinsic T\sqrt{T} parametric bandit complexity.

  • Systematic numerical experiments on both simulated and real datasets are carried out, supporting our theoretical findings, and highlighting their practicality.

The rest of the paper is organised as follows. § 2 introduces the problem formulation and the Lipschitz fairness constraint. § 3 characterises the structural properties of optimal fair policies, including the cost of fairness and the canonical boundary representation. § 4 develops fair learning algorithms and establishes their regret upper bounds under continuous and general context distributions. § 5 derives matching minimax lower bounds. § 6 presents numerical experiments on synthetic and real datasets. § 7 concludes. Additional algorithms, technical proofs and supplementary numerical results are provided in the Appendix.

Notation. For any positive integer mm, denote [m]={1,…,m}[m]=\{1,\ldots,m\}. For a∈Ra\in\mathbb{R}, let ⌈a⌉\lceil a\rceil and ⌊a⌋\lfloor a\rfloor, and let [a][0,1]=min⁡{1,max⁡{0,a}}[a]_{[0,1]}=\min\{1,\max\{0,a\}\} denote the clipping of aa onto [0,1][0,1]. For a,b∈Ra,b\in\mathbb{R}, write a∨b=max⁡{a,b}a\vee b=\max\{a,b\} and a∧b=min⁡{a,b}a\wedge b=\min\{a,b\}, and let {a}+=a∨0\{a\}_{+}=a\vee0. For ν≥1\nu\ge1 and any vector v∈Rpv\in\mathbb{R}^{p}, let ∥v∥ν\lVert v\rVert_{\nu} denote its ℓν\ell_{\nu}-norm, and let ∥v∥ν∗\lVert v\rVert_{\nu^{*}} denote its dual norm, where 1/ν+1/ν∗=11/\nu+1/\nu^{*}=1, with the convention 1/∞=01/\infty=0. For a point xx and any sets AA and BB, write dist⁡ν(x,A)=inf⁡z∈A∥x−z∥ν\operatorname{dist}_{\nu}(x,A)=\inf_{z\in A}\lVert x-z\rVert_{\nu}, dist⁡ν(A,B)=inf⁡y∈A,z∈B∥y−z∥ν\operatorname{dist}_{\nu}(A,B)=\inf_{y\in A,z\in B}\lVert y-z\rVert_{\nu}, and diam⁡ν(A)=sup⁡x,z∈A∥x−z∥ν\operatorname{diam}_{\nu}(A)=\sup_{x,z\in A}\lVert x-z\rVert_{\nu}. For any two symmetric matrices AA and BB, write A⪰BA\succeq B when A−BA-B is positive semidefinite, and let λmin⁡(A)\lambda_{\min}(A) be the smallest eigenvalue of AA. For any function f:X→Rf:\mathcal{X}\to\mathbb{R} and a subset A⊆XA\subseteq\mathcal{X}, denote by f∣A:A→Rf|_{A}:A\to\mathbb{R} the restriction of ff to AA. For two sequences of positive numbers {an}\{a_n\} and {bn}\{b_n\}, write an≲bna_n\lesssim b_n if there exist a universal constant C>0C>0 and some n0∈Nn_0\in\mathbb{N} such that an≤Cbna_n\le Cb_n for all n≥n0n\ge n_0, and write an≍bna_n\asymp b_n if an≲bna_n\lesssim b_n and an≳bna_n\gtrsim b_n. We further write an≲log⁡bna_n\lesssim_{\log} b_n (or an=Olog⁡(bn)a_n=O_{\log}(b_n)) if an≲bna_n\lesssim b_n holds up to poly-logarithmic factors. For random variables {Xn}\{X_n\} and positive numbers {an}\{a_n\}, write Xn=Op(an)X_n=O_p(a_n) if lim⁡M→∞lim sup⁡n→∞P(∣Xn∣≥Man)=0\lim_{M\to\infty}\limsup_{n\to\infty}\mathbb{P}(|X_n|\ge Ma_n)=0. For any probability measure PP, let ∥f∥L2(P)={EP(f2)}1/2\lVert f\rVert_{L_2(P)}=\{\mathbb{E}_P(f^2)\}^{1/2}. We write λd(⋅)\lambda_d(\cdot) for dd-dimensional Lebesgue measure.

Problem formulation

In this section, we formalise the problem setup for Lipschitz fair contextual bandits in Section 2.1, followed by technical assumptions in § 2.2.

Setup and (L,ν)(L,\nu)-Lipschitz fairness

Consider a two-armed contextual bandit problem over a finite time horizon T∈Z+T \in\mathbb{Z}_{+}. The two-arm setting is widely studied and arises naturally in applications such as clinical trials, where the decision involves whether to administer a particular treatment. We focus on this setting for simplicity of exposition and to highlight the key ideas of our framework, with discussions on the extensions to multiple arms (K≥3K \ge3) in § 7. At each round t∈[T]t \in[T], a context Xt∈X=[0,1]dX_t \in\mathcal{X} = [0,1]^d is observed before an action is chosen, with both arms sharing the same context. After observing XtX_t, the learner chooses an action At∈[2]A_t \in[2] and receives the realised reward Yt=Rt,AtY_t = R_{t,A_t} of the selected arm.

For x∈[0,1]dx \in[0,1]^d, write x~=(1,x⊤)⊤∈Rd+1\widetilde{x} = (1, x^\top)^\top\in\mathbb{R}^{d+1}. We assume the linear reward model Rt,k=X~t⊤βk∗+ξt,kR_{t,k} = \widetilde{X}_t^\top\beta_k^* + \xi_{t,k}, k∈[2]k \in[2], where βk∗∈Rd+1\beta_k^* \in\mathbb{R}^{d+1} is an unknown arm-specific parameter and ξt,k\xi_{t,k} is a mean-zero noise variable. Write θ∗=β1∗−β2∗=(a∗,γ∗⊤)⊤\theta^* = \beta_1^* - \beta_2^* = (a^*, \gamma^{*\top})^\top, where a∗∈Ra^* \in\mathbb{R} and γ∗∈Rd\gamma^* \in\mathbb{R}^d denote the intercept and slope differences, respectively.

Remark 1. Our framework also extends to generalised linear rewards Rt,k=μ(X~t⊤βk∗)+ξt,kR_{t,k} = \mu(\widetilde{X}_t^\top\beta_k^*) + \xi_{t,k}, k∈[2]k \in[2], with a known, strictly increasing inverse link function μ\mu, and mean-zero noise ξt,k\xi_{t,k} (e.g. [22] (Li et al. 2017)). Monotonicity of μ\mu preserves the sign of the reward gap and consequently the linear decision boundary, allowing the structural characterisation of fair policies discussed in § 3 to extend under suitable regularity conditions.

A policy is modelled as a measurable function p(⋅):[0,1]d→[0,1]p(\cdot) : [0,1]^d \to[0,1], where p(x)p(x) denotes the probability of pulling arm 1 at context xx, i.e. p(x)=P(At=1∣X=x)p(x) = \mathbb{P}(A_t = 1 \mid X = x). In this setting, at the population level, the standard reward-maximisation objective for bandits is to seek a policy pp that maximises the expected reward

R(p)=E[X~⊤β1∗p(X)+X~⊤β2∗{1−p(X)}]=E[X~⊤β2∗+Δ∗(X)p(X)],(1)R(p) = \mathbb{E}\left[\widetilde{X}^\top\beta_1^*p(X) + \widetilde{X}^\top\beta_2^*\{1-p(X)\}\right] = \mathbb{E}\left[\widetilde{X}^\top\beta_2^* + \Delta^*(X)p(X)\right], \tag*{(1)}

References

  1. [1]Abbasi-Yadkori, Y., Pál, D., and Szepesvári, C. (2011). Improved algorithms for linear stochastic bandits. Advances in neural information processing systems, 24.
  2. [2]Ali, M., Sapiezynski, P., Bogen, M., Korolova, A., Mislove, A., and Rieke, A. (2019). Discrimination through optimization: How facebook’s ad delivery can lead to biased outcomes. Proceedings of the ACM on human-computer interaction, 3(CSCW):1–30.
  3. [3]Ameko, M. K., Beltzer, M. L., Cai, L., Boukhechba, M., Teachman, B. A., and Barnes, L. E. (2020). Offline contextual multi-armed bandits for mobile health interventions: A case study on emotion regulation. In Proceedings of the 14th ACM Conference on Recommender Systems, pages 249–258.DOI
  4. [4]Barman, S., Khan, A., Maiti, A., and Sawarni, A. (2023). Fairness and welfare quantification for regret in multi-armed bandits. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol 37, No 6, pages 6762–6769.arxiv.org/abs/2205.13930
  5. [5]Bastani, H., Bayati, M., and Khosravi, K. (2021). Mostly exploration-free algorithms for contextual bandits. Management Science, 67(3):1329–1349.arxiv.org/abs/1704.09011
  6. [6]Becker, B. and Kohavi, R. (1996). Adult. UCI Machine Learning Repository.
  7. [7]Blanchard, M. and Goyal, V. (2025). Collaborative min-max regret in grouped multi-armed bandits. arXiv preprint arXiv:2506.10313.arxiv.org/abs/2506.10313
  8. [8]Bousquet, O. (2002). A bennett concentration inequality and its application to suprema of empirical processes. Comptes Rendus Mathematique, 334(6):495–500.DOI
  9. [9]Cannelli, L., Nuti, G., Sala, M., and Szehr, O. (2023). Hedging using reinforcement learning: Contextual k-armed bandit versus q-learning. The Journal of Finance and Data Science, 9:100101.DOI
  10. [10]Cesa-Bianchi, N., Gaillard, P., Gentile, C., and Gerchinovitz, S. (2017). Algorithmic chaining and the role of partial feedback in online nonparametric learning. In Conference on Learning Theory, pages 465–481. PMLR.
  11. [11]Chaudhary, S. and Sinha, A. (2023). α-fair contextual bandits. arXiv preprint arXiv:2310.14164.arxiv.org/abs/2310.14164
  12. [12]Chen, G., Li, X., and Ye, Y. (2021). Fairer LP-based online allocation via analytic center. arXiv preprint arXiv:2110.14621.arxiv.org/abs/2110.14621
  13. [13]Chen, X., Lyu, J., Zhang, X., and Zhou, Y. (2026a). Technical note—fairness-aware online price discrimination with nonparametric demand models. Operations Research, 74(1):118–129.DOI
  14. [14]Chen, X., Simchi-Levi, D., and Wang, Y. (2022). Privacy-preserving dynamic personalized pricing with demand learning. Management Science, 68(7):4878–4898.
  15. [15]Chen, X., Simchi-Levi, D., and Wang, Y. (2026b). Utility fairness in contextual dynamic pricing with demand learning. Management Science, 72(3):2619–2633.
  16. [16]Chen, Y., Cuellar, A., Luo, H., Modi, J., Nemlekar, H., and Nikolaidis, S. (2020). Fair contextual multi-armed bandits: Theory and experiments. In Conference on Uncertainty in Artificial Intelligence, pages 181–190. PMLR.arxiv.org/abs/1912.08055
  17. [17]Chu, W., Li, L., Reyzin, L., and Schapire, R. (2011). Contextual bandits with linear payoff functions. In Proceedings of the fourteenth international conference on artificial intelligence and statistics, pages 208–214. JMLR Workshop and Conference Proceedings.
  18. [18]Cohen, M. C., Miao, S., and Wang, Y. (2025). Dynamic pricing with fairness constraints. Operations Research, 73(6):3027–3043.
  19. [19]Cyberspace Administration of China et al. (2022). Provisions on the administration of algorithmic recommendation for internet information services. Article 21.
  20. [20]Dani, V., Hayes, T. P., and Kakade, S. M. (2008). Stochastic linear optimization under bandit feedback. In Proceedings of the 21st Annual Conference on Learning Theory, pages 355–366.
  21. [21]Dwork, C., Hardt, M., Pitassi, T., Reingold, O., and Zemel, R. (2012). Fairness through awareness. In Proceedings of the 3rd Innovations in Theoretical Computer Science Conference, pages 214–226. Association for Computing Machinery.
  22. [22]Filippi, S., Cappe, O., Garivier, A., and Szepesvári, C. (2010). Parametric bandits: The generalized linear case. Advances in neural information processing systems, 23.
  23. [23]Fukuchi, K. and Sakuma, J. (2023). Demographic parity constrained minimax optimal regression under linear model. In Advances in Neural Information Processing Systems, volume 36, pages 8653–8689.arxiv.org/abs/2206.11546
  24. [24]Gillen, S., Jung, C., Kearns, M., and Roth, A. (2018). Online learning with an unknown fairness metric. Advances in neural information processing systems, 31.arxiv.org/abs/1802.06936
  25. [25]Goldenshluger, A. and Zeevi, A. (2013). A linear response bandit problem. Stochastic Systems, 3(1):230–261.DOI
  26. [26]Grazzi, R., Akhavan, A., Falk, J. I., Cella, L., and Pontil, M. (2022). Group meritocratic fairness in linear contextual bandits. Advances in Neural Information Processing Systems, 35:24392–24404.arxiv.org/abs/2206.03150
  27. [27]Gupta, S. and Kamble, V. (2021). Individual fairness in hindsight. Journal of Machine Learning Research, 22(144):1–35.
  28. [28]Harville, D. A. (1998). Matrix algebra from a statistician’s perspective.
  29. [29]Hou, X. and Zhang, L. (2024). Finite-sample and distribution-free fair classification: Optimal trade-off between excess risk and fairness, and the cost of group-blindness. arXiv preprint arXiv:2410.16477.arxiv.org/abs/2410.16477
  30. [30]Jolliffe, I. T. and Cadima, J. (2016). Principal component analysis: a review and recent developments. Philosophical transactions. Series A, Mathematical, physical, and engineering sciences, 374(2065):20150202.DOI
  31. [31]Joseph, M., Kearns, M., Morgenstern, J., Neel, S., and Roth, A. (2018). Meritocratic fairness for infinite and contextual bandits. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, pages 158–163.DOI
  32. [32]Joseph, M., Kearns, M., Morgenstern, J. H., and Roth, A. (2016). Fairness in learning: Classic and contextual bandits. Advances in neural information processing systems, 29.arxiv.org/abs/1605.07139
  33. [33]Lattimore, T. and Szepesvári, C. (2020). Bandit Algorithms. Cambridge University Press.
  34. [34]Li, F., Liu, J., and Ji, B. (2019). Combinatorial sleeping bandits with fairness constraints. IEEE Transactions on Network Science and Engineering, 7(3):1799–1813.
  35. [36]Liu, Q., Xu, W., Wang, S., and Fang, Z. (2022). Combinatorial bandits with linear constraints: Beyond knapsacks and fairness. Advances in Neural Information Processing Systems, 35:2997–3010.DOI
  36. [37]Liu, Y., Radanovic, G., Dimitrakakis, C., Mandal, D., and Parkes, D. C. (2017). Calibrated fairness in bandits. arXiv preprint arXiv:1707.01875.arxiv.org/abs/1707.01875
  37. [39]Mao, J., Leme, R., and Schneider, J. (2018). Contextual pricing for lipschitz buyers. Advances in Neural Information Processing Systems, 31.
  38. [40]McShane, E. J. (1934). Extension of range of functions. Bulletin of the American Mathematical Society, 40(12):837 – 842.DOI
  39. [41]National Institute of Standards and Technology (2023). Artificial intelligence risk management framework (AI RMF 1.0). https://www.nist.gov/itl/ai-risk-management-framework.
  40. [42]Patil, V., Ghalme, G., Nair, V., and Narahari, Y. (2021). Achieving fairness in the stochastic multi-armed bandit problem. Journal of Machine Learning Research, 22(174):1–31.arxiv.org/abs/1907.10516
  41. [43]Perchet, V. and Rigollet, P. (2013). The multi-armed bandit problem with covariates. The Annals of Statistics, 41(2):693–721.
  42. [45]Salem, J., Gupta, S., and Kamble, V. (2026). Algorithmic challenges in ensuring fairness at the time of decision. Operations Research, 74(2):1005–1025.arxiv.org/abs/2103.09287
  43. [46]Sawarni, A., Pal, S., and Barman, S. (2023). Nash regret guarantees for linear bandits. Advances in Neural Information Processing Systems, 36:333288–33318.arxiv.org/abs/2310.02023
  44. [47]Tankha, A. (2023). 70k+ Job Applicants Data (Human Resource). Kaggle.
  45. [48]Tropp, J. A. (2015). An introduction to matrix concentration inequalities. Foundations and trends® in machine learning, 8(1-2):1–230.arxiv.org/abs/1501.01571
  46. [49]Van Der Vaart, A. W. and Wellner, J. A. (1996). Weak convergence. In Weak convergence and empirical processes: with applications to statistics, pages 16–28. Springer.DOI
  47. [50]Vershynin, R. (2018). High-dimensional probability: An introduction with applications in data science, volume 47. Cambridge university press.
  48. [51]Wainwright, M. J. (2019). High-dimensional statistics: A non-asymptotic viewpoint, volume 48. Cambridge university press.
  49. [52]Wang, L., Bai, Y., Sun, W., and Joachims, T. (2021). Fairness of exposure in stochastic bandits. In International Conference on Machine Learning, pages 10686–10696. PMLR.arxiv.org/abs/2103.02735
  50. [53]Wei, W., Ma, X., and Wang, J. (2023). Fair adaptive experiments. Advances in Neural Information Processing Systems, 36:19157–19169.arxiv.org/abs/2310.16290
  51. [55]Wu, X. and Li, B. (2024). Achieving regular and fair learning in combinatorial multi-armed bandit. In IEEE INFOCOM 2024-IEEE Conference on Computer Communications, pages 361–370. IEEE.DOI
  52. [57]Yu, B. (1997). Assouad, fano, and le cam. In Festschrift for Lucien Le Cam: research papers in probability and statistics, pages 423–435. Springer.DOI
  53. [58]Zeng, X., Jiang, K., Cheng, G., and Dobriban, E. (2026). Bayes-optimal fair classification with linear disparity constraints via pre-, in-, and post-processing. Journal of Machine Learning Research, 27(65):1–87.
  54. [59]Zhu, Z. and Van Roy, B. (2023). Scalable neural contextual bandit for recommender systems. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, pages 3636–3646.arxiv.org/abs/2306.14834

Paper details

vlnc:W_k7qegw49
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents