Pith. sign in

REVIEW 3 major objections 3 minor 35 references

Improved Online Confidence Bounds for Multinomial Logistic Bandits

T0 review · 3 major / 3 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper claims that online confidence bounds for multinomial logistic choice models can be made independent of the parameter norm bound B and the maximum assortment size K, and that the resulting bounds yield variance-dependent…

desk verdict Real technical advances in MNL bandit confidence bounds, but the headline B-free variance-dependent regret is only established under an unstated cumulative-variance condition. read the letter →

arxiv 2502.10020 v5 pith:DKIJPXFJ submitted 2025-02-14 stat.ML cs.LG

Joongkyu Lee , Min-hwan Oh This is my paper · ORCID
classification stat.MLcs.LG
keywords multinomiallogisticbanditsonlineconfidenceboundsvariance-dependentregretself-concordantfunctionsVille'sinequalityassortmentoptimizationmirrordescentmaximumlikelihoodestimation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper sets out to prove that the confidence radius for online estimation in multinomial logistic (MNL) choice models can be reduced from $O(B\sqrt{d}\log t\log K)$ to $O(\sqrt{d\log t}+B\sqrt{d})$, and eventually to $O(\sqrt{d\log t})$ at large horizons, with no dependence on the norm bound $B$ or the maximum assortment size $K$. It claims this tighter radius, enforced by updating the parameter only inside a shrinking ellipsoid, is enough to give contextual MNL bandits their first variance-dependent regret bounds: $O(d\log T\sqrt{\sum_{t=1}^T\sigma_t^2})$ for a constant-time algorithm (OFU-MNL++) when $T$ is large, and $O(d\log(BT)\sqrt{\sum_{t=1}^T\sigma_t^2})$ for an MLE-based algorithm (OFU-M2NL) at every horizon. If these claims hold, assortment-selection systems would stop paying penalties that grow exponentially in the parameter scale or linearly in assortment size, and their regret would adapt to how noisy observed rewards are rather than scaling with worst-case $T$. The technical engine is a self-concordant-like bound on the MNL loss in the $\ell^\infty$ norm plus a martingale argument that replaces a smoothing step costing $\log K$.

What carries the argument

The machinery has three load-bearing pieces. First, the paper proves the MNL loss is $3\sqrt{2}$-self-concordant-like in the $\ell^\infty$ norm, meaning third derivatives of the loss along a direction are controlled by the Hessian times the $\ell^\infty$ norm of the direction; this lets the online step size $\eta$ depend on the update-condition constant $\alpha$ rather than on $B$ and $\log K$. Second, it forms a supermartingale $A_t=\exp(\sum_s(\ell_s(w^*)-\bar\ell_s(\tilde z_s)))$ over an intermediary softmax parameter $\tilde z_s$, so Ville's inequality bounds the loss-gap term by $\log(1/\delta)$ instead of the $O(\log K(\log t)^2)$ incurred by the smoothed Bernstein argument. Third, the adaptive warm-up phase builds an ellipsoid $W_t^w(\delta)$ around an auxiliary online estimate and refuses to plan on any context whose $H_w$-weighted norm is large, keeping $\alpha=1/(3\sqrt{2})$ during planning; the subsequent regret proof decomposes the per-round regret into a covariance between rewards and the $H_t^{-1}$ norm of context vectors and applies an elliptical-potential bound to that covariance.

What would settle it

Track, in a simulation with adversarially correlated contexts, whether the true parameter $w^*$ stays inside the warm-up ellipsoid $W_t^w(\delta)$ on every round; if the containment frequency drops below $1-\delta$ or the empirical confidence radius exceeds $\beta_t(\delta)$, the constant-$\alpha$ condition that sustains the $O(\sqrt{d\log t})$ radius has failed, and the regret bound reverts to the $B$-dependent warm-up radius.

Watch

Extended reading notes

Core claim

The central claim is Theorem 4.2: when the online mirror-descent update is restricted to a set where the deviation $|x_{ti}^\top(w-w^*)|$ is at most $\alpha$, the estimation error satisfies $\|w_t-w^*\|_{H_t}=O(\alpha\sqrt{d\log(t/\delta)}+B\sqrt{\lambda})$ with high probability. Taking $\alpha$ constant (via the warm-up ellipsoid) and $\lambda=144d$ turns this into $O(\sqrt{d\log t}+B\sqrt{d})$, and for $t\ge e^{B^2}$ the $B\sqrt{d}$ term is absorbed. From this the paper derives, under uniform and non-uniform rewards, regret bounds proportional to $\sqrt{\sum_t \sigma_t^2}$, the square root of cumulative reward variance, where earlier bounds scaled as $B^{3/2}d\log K(\log T)^{3/2}\sqrt{T}$.

Load-bearing premise

The whole improvement rests on the true parameter $w^*$ remaining inside the warm-up confidence ellipsoid at every round; if that containment ever fails, the update-condition constant $\alpha$ can no longer be held small, and the confidence radius reverts to the $B$-dependent warm-up scale.

Editorial extensions

If this is right

  • If Theorem 4.5 is correct, then for horizons $T\ge \tilde O(e^{B^2}+B^8d^2/\kappa^2)$ the OFU-MNL++ algorithm achieves regret $O(d\log T\sqrt{\sum_{t=1}^T\sigma_t^2})$ with per-round computational cost $O(Kd^3+\mathrm{poly}(N))$, independent of $t$.
  • If Theorem 4.12 is correct, OFU-M2NL achieves $O(d\log(BT)\sqrt{\sum_{t=1}^T\sigma_t^2})$ regret for every $T$, with no $B$ or $K$ factor in the leading term, at the price of per-round cost that grows linearly with $t$.
  • Under uniform rewards ($r_{ti}=1$), Proposition 4.10 gives instance-dependent regret $\tilde O(d\sqrt{\sum_t \kappa_t^*})$, improving the earlier $e^B$ factor.
  • In the worst case over rewards, $\sigma_t\le 1$ makes the OFU-MNL++ bound $\tilde O(d\sqrt{T})$, matching the $\Omega(d\sqrt{T})$ minimax lower bound from the paper.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper, the same $\ell^\infty$-self-concordance route should transfer to other softmax-parameterized losses, such as top-$k$ or hierarchical softmax, where the number of categories enters only through a log factor that the Ville-inequality step would remove.
  • Beyond the paper, the empirical tuning of $\tau_t$ in the experiments suggests the theoretical warm-up threshold is conservative; an adaptive threshold that lowers $\tau$ once the elliptical potential stabilizes could shorten the $e^{B^2}$ burn-in without sacrificing the $B$-free radius.
  • Beyond the paper, the variance-dependent form of the bound suggests a practical diagnostic: in settings where observed reward variance is low, the planning-phase updates dominate and the algorithm should approach its asymptotic regret quickly; logging per-round $\sigma_t^2$ alongside regret would test this directly.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. The paper develops an improved online confidence bound for multinomial logistic (MNL) models, based on new self-concordance-like properties in the ℓ∞-norm and on Ville's inequality rather than the previously used Bernstein-type arguments. This bound is then applied to two MNL bandit algorithms: OFU-MNL++, a constant-time algorithm with an adaptive warm-up phase and a restricted confidence ellipsoid, and OFU-M2NL, an MLE-based algorithm. The paper claims regret bounds of the form O((d log T + B d sqrt(log T)) sqrt(sum_t σ_t^2)) plus additive 1/κ terms for OFU-MNL++, and O(d log(BT) sqrt(sum_t σ_t^2)) plus additive 1/κ terms for OFU-M2NL, and further claims that the additive terms vanish for sufficiently large T, yielding B-free, K-free variance-dependent regret.

Significance. The technical core of the paper is valuable: the ℓ∞-norm self-concordance properties in Propositions B.3–B.6 are a genuine improvement over the existing ℓ2-norm treatments, and the Ville-based supermartingale argument in Lemma C.2 gives a cleaner bound on the main estimation term than the smoothed Bernstein argument of Lee and Oh (2024). The explicit theorem statements with additive 1/κ terms are plausible and would already be a useful contribution to the MNL bandit literature, especially the constant-time OFU-MNL++ update and the variance-dependent regret decomposition. However, the headline asymptotic claim that the regret becomes B-free and K-free for sufficiently large T is not established by the stated theorems for low-variance instances, and the MLE-based 'poly(B),K-free anytime' claim is also overstated. No code is provided, so reproducibility rests on the mathematical appendices.

major comments (3)
  1. [Appendix D.2.3, Lemma D.6] The claimed simplification of Theorem 4.5 to O(d log T sqrt(sum_{t=1}^T σ_t^2)) for T ≥ Õ(e^{B^2} + B^8 d^2 / κ^2) is not a consequence of the stated bound. Theorem 4.5 gives Reg_T ≤ O((d log T + B d sqrt(log T)) sqrt(sum σ_t^2) + (1/κ)B^3 d^2 (log T)^2 + (1/κ)B^4 d log T). The threshold T ≥ Õ(e^{B^2} + B^8 d^2/κ^2) only ensures that the B d sqrt(log T) factor in the leading term is dominated by d log T. Dominating the additive terms requires sqrt(sum σ_t^2) ≥ (1/κ)B^3 d log T + (1/κ)B^4, and no such lower bound on the reward variance is assumed or implied. For example, if σ_t^2 = 1/T for all t, then sqrt(sum σ_t^2) = 1, and at T = e^{B^2} the additive term is about (1/κ)B^7 d^2 while the variance term is d B^2; since 1/κ = O(K^2 e^{3B}), the additive term is exponentially larger. The explicit two-term bound in Theorem 4.5 may be correct, but the advertised B-free, K-free regime requires either an explicit lower-bound condition on sum σ_t^2 or a reformulation of the claim.
  2. [Appendix D.2.3, Lemma D.6] The proof of Lemma D.6 is not valid as written. It argues that if r_ti < tR_t(S_t), then removing item i from S_t would increase the optimistic revenue tR_t(S_t), contradicting optimality. But removing i increases tR_t(S_t) if and only if r_ti < tR_t(S_t \ {i}), and tR_t(S_t \ {i}) can be larger than tR_t(S_t). The lemma itself is true, because for an optimal assortment under an MNL model the revenue of the assortment is a weighted average of r_ti and the revenue after removing i, so r_ti ≥ tR_t(S_t) follows from optimality, but the displayed contradiction needs a repaired argument.
  3. [Section 4.3, Theorem 4.12 and discussion] The claim that OFU-M2NL achieves 'completely poly(B), K-free regret' is too strong. The theorem states a regret bound of O(d log(BT) sqrt(sum σ_t^2) + (1/κ)d^2(log(BT))^2), and 1/κ = O(K^2 e^{3B}) is exponential in B and quadratic in K. Thus the additive term is neither poly(B)-free nor K-free. Moreover, the same low-variance issue as in Theorem 4.5 applies: the additive term can dominate the variance term unless an explicit lower bound on sqrt(sum σ_t^2) is imposed. The leading-term claim with an explicit additive term would be fine, but the wording in the abstract and in the discussion overstates what the theorem proves.
minor comments (3)
  1. [Section 5 and Appendix G] The experiments tune the threshold τ_t manually rather than using the theoretical value τ_t = 6√2 ζ_t(δ) specified in Algorithm 1. This is a sensible practical choice, but it should be stated explicitly in the main text as an implementation detail, not described only in the appendix.
  2. [Theorem 4.2 and Lemma C.1] The step-size parameter η appears in slightly different algebraic forms in Theorem 4.2 and in Lemma C.1 in Appendix C. If this is not a typographical artifact, the two statements should be harmonized and the proof of Lemma C.1 should be checked against the exact value used in the theorem.
  3. [Appendix F] Lemma 4.11 is stated in the main text as informal, and the formal version correctly relies on Theorem 3.1 of Lee et al. (2024b). Please make the dependence on that external result explicit at the point where the confidence radius is defined in the main text, rather than only in the appendix.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the restricted-space confidence bound is a genuine two-stage bootstrap, not a fitted or self-referential prediction.

full rationale

The derivation chain is self-contained at the level of the main theorems. Theorem 4.2 is proved in Appendix C from the new l8-self-concordant properties (Propositions B.3-B.6), a supermartingale argument via Ville's inequality (Lemma C.2), and self-concordant quadratic approximation lemmas (Lemmas C.3, C.5, C.6); it does not assume the constant-alpha condition. Corollary D.1 obtains the warm-up radius zeta_t by applying Theorem 4.2 with W_t = W and alpha = 2B, which is valid because the full space W gives alpha = 2B by Cauchy-Schwarz. Corollary D.2 then uses w* in W_t^w(delta) and tau_t = 6*sqrt(2)*zeta_t(delta) to verify the Theorem 4.2 update condition with alpha = 1/(3*sqrt(2)), and applies Theorem 4.2 a second time. This is a legitimate two-stage argument, not a circular reduction: the planning-phase bound is conditional on a high-probability event that is itself proved rather than assumed. Theorem 4.5's regret decomposition (D.5)-(D.10) is an independent argument, and sigma_t^2 is an instance-dependent quantity defined from the model, not a fitted parameter renamed as a prediction. Citations to Lee & Oh (2024) and Lee et al. (2024b) are auxiliary published lemmas with proofs and stated assumptions (e.g., Lemmas D.3, D.5, D.9, D.11 and Lemma F.1); they do not smuggle in the target B-free/K-free conclusion. The discussion's simplification at T >= O~(e^{B^2} + B^8 d^2/kappa^2) may overclaim mathematically relative to the additive terms, but that is a quantitative correctness concern, not an equivalence-by-construction or self-citation-driven circularity.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The theoretical results introduce no fitted parameters: λ, η, β, ζ are given as explicit functions of d, B, δ. The only fitted quantity is the experimental threshold τ_t in Section G. The proof rests on standard boundedness (Assumption 3.1), the κ-constant (Definition 3.2), correctness of the MNL choice model, and the imported GLM confidence sequence (Lee et al. 2024b). No new entities (forces, particles, conserved quantities) are postulated.

free parameters (1)
  • τ_t (warm-up threshold in experiments) = tuned over an unreported range (Appendix G)
    The threshold τ_t is set as 6√2 ζ_t(δ) in theory, but in the numerical experiments it is 'empirically tuned over a certain range of values'. The chosen values are not reported, so the empirical regret curves are not fully reproducible from the paper alone.
assumptions (4)
  • domain assumption Assumption 3.1: ||x_ti||₂ ≤ 1, r_ti ∈ [0,1], and ||w*||₂ ≤ B for known B.
    Standard boundedness in the MNL bandit literature; used throughout the proofs and to set λ and η.
  • domain assumption Definition 3.2: existence of κ > 0 with min pt(i)p_t(0) ≥ κ over all feasible assortments and w ∈ W.
    Needed to bound the number of warm-up rounds (Lemma D.10) and the additive regret terms; κ can be exponentially small, O(K²e^{3B}), which drives the 'sufficiently large T' condition.
  • standard math Theorem 3.1 of Lee et al. (2024b) (unified confidence sequence for GLMs) is valid for the MNL loss as used in Lemma 4.11.
    The MLE-based bound imports this published result as a black box. The paper does not verify the conditions (e.g., exp-concavity, vector-response extension) for the multinomial logistic likelihood.
  • domain assumption The MNL choice model is correctly specified with linear utility x^T w* and known feature set each round.
    The entire regret analysis assumes the choice probabilities follow the MNL model; misspecification is out of scope.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Improved Online Confidence Bounds for Multinomial Logistic Bandits." pith.science (2026). https://pith.science/paper/DKIJPXFJ

@misc{pith2026250210020,
  author       = {Pith},
  title        = {Pith review of: Improved Online Confidence Bounds for Multinomial Logistic Bandits},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DKIJPXFJ}},
  note         = {Machine review of arXiv:2502.10020}
}
abstract

In this paper, we propose an improved online confidence bound for multinomial logistic (MNL) models and apply this result to MNL bandits, achieving variance-dependent optimal regret. Recently, Lee & Oh (2024) established an online confidence bound for MNL models and achieved nearly minimax-optimal regret in MNL bandits. However, their results still depend on the norm-boundedness of the unknown parameter $B$ and the maximum size of possible outcomes $K$. To address this, we first derive an online confidence bound of $O\left(\sqrt{d \log t} + B \sqrt{d} \right)$, which is a significant improvement over the previous bound of $O (B \sqrt{d} \log t \log K )$ (Lee & Oh, 2024). This is mainly achieved by establishing tighter self-concordant properties of the MNL loss and applying Ville's inequality to bound the estimation error. Using this new online confidence bound, we propose a constant-time algorithm, OFU-MNL++, which achieves a variance-dependent regret bound of $O \Big( d \log T \sqrt{ \sum_{t=1}^T \sigma_t^2 } \Big) $ for sufficiently large $T$, where $\sigma_t^2$ denotes the variance of the rewards at round $t$, $d$ is the dimension of the contexts, and $T$ is the total number of rounds. Furthermore, we introduce a Maximum Likelihood Estimation (MLE)-based algorithm, OFU-MN$^2$L, which achieves an anytime poly(B)-free regret of $O \Big( d \log (BT) \sqrt{ \sum_{t=1}^T \sigma_t^2 } \Big) $.

Figures

Figures reproduced from arXiv: 2502.10020 by the authors.

Figure 1
Figure 1. Cumulative regret for varying the norm-boundedness of the unknown parameter [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Runtime per round for varying the norm-boundedness of the unknown parameter [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 14 canonical work pages

  1. [1]

    Instance-wise minimax-optimal algorithms for logistic bandits

    Abeille, M., Faury, L., and Calauz \`e nes, C. Instance-wise minimax-optimal algorithms for logistic bandits. In International Conference on Artificial Intelligence and Statistics, pp.\ 3691--3699. PMLR, 2021

    2021

  2. [2]

    A tractable online learning algorithm for the multinomial logit contextual bandit

    Agrawal, P., Tulabandhula, T., and Avadhanula, V. A tractable online learning algorithm for the multinomial logit contextual bandit. European Journal of Operational Research, 310 0 (2): 0 737--750, 2023

    2023

  3. [3]

    Thompson sampling for the mnl-bandit

    Agrawal, S., Avadhanula, V., Goyal, V., and Zeevi, A. Thompson sampling for the mnl-bandit. In Conference on learning theory, pp.\ 76--78. PMLR, 2017

    2017

  4. [4]

    Mnl-bandit: A dynamic learning approach to assortment selection

    Agrawal, S., Avadhanula, V., Goyal, V., and Zeevi, A. Mnl-bandit: A dynamic learning approach to assortment selection. Operations Research, 67 0 (5): 0 1453--1485, 2019

    2019

  5. [5]

    and Orabona, F

    Campolongo, N. and Orabona, F. Temporal variability in implicit online learning. Advances in neural information processing systems, 33: 0 12377--12387, 2020

    work page 2020

  6. [6]

    Dynamic assortment optimization with changing contextual information

    Chen, X., Wang, Y., and Zhou, Y. Dynamic assortment optimization with changing contextual information. The Journal of Machine Learning Research, 21 0 (1): 0 8918--8961, 2020

    work page 2020

  7. [7]

    Randomized Exploration for Reinforcement Learning with Multinomial Logistic Function Approximation

    Cho, W., Hwang, T., Lee, J., and Oh, M.-h. Randomized exploration for reinforcement learning with multinomial logistic function approximation. arXiv preprint arXiv:2405.20165, 2024

    work page Pith review arXiv 2024

  8. [8]

    M., Gallego, G., and Topaloglu, H

    Davis, J. M., Gallego, G., and Topaloglu, H. Assortment optimization under variants of the nested logit model. Operations Research, 62 0 (2): 0 250--273, 2014

    work page 2014

Show all 35 references
  1. [9]

    On the performance of thompson sampling on logistic bandits

    Dong, S., Ma, T., and Van Roy, B. On the performance of thompson sampling on logistic bandits. In Conference on Learning Theory, pp.\ 1158--1160. PMLR, 2019

    2019

  2. [10]

    Improved optimistic algorithms for logistic bandits

    Faury, L., Abeille, M., Calauz \`e nes, C., and Fercoq, O. Improved optimistic algorithms for logistic bandits. In International Conference on Machine Learning, pp.\ 3052--3060. PMLR, 2020

    2020

  3. [11]

    Jointly efficient and optimal algorithms for logistic bandits

    Faury, L., Abeille, M., Jun, K.-S., and Calauz \`e nes, C. Jointly efficient and optimal algorithms for logistic bandits. In International Conference on Artificial Intelligence and Statistics, pp.\ 546--580. PMLR, 2022

    2022

  4. [12]

    J., Kale, S., Luo, H., Mohri, M., and Sridharan, K

    Foster, D. J., Kale, S., Luo, H., Mohri, M., and Sridharan, K. Logistic regression: The importance of being improper. In Conference on learning theory, pp.\ 167--208. PMLR, 2018

    2018

  5. [13]

    Hazan, E. et al. Introduction to online convex optimization. Foundations and Trends in Optimization , 2 0 (3-4): 0 157--325, 2016

    2016

  6. [14]

    and Oh, M.-h

    Hwang, T. and Oh, M.-h. Model-based reinforcement learning with multinomial logistic function approximation. arXiv preprint arXiv:2212.13540, 2022

    2022 arXiv

  7. [15]

    Mixability made efficient: Fast online multiclass logistic regression

    J \'e z \'e quel, R., Gaillard, P., and Rudi, A. Mixability made efficient: Fast online multiclass logistic regression. Advances in Neural Information Processing Systems, 34: 0 23692--23702, 2021

    2021

  8. [16]

    Jin, C., Yang, Z., Wang, Z., and Jordan, M. I. Provably efficient reinforcement learning with linear function approximation. In Conference on Learning Theory, pp.\ 2137--2143. PMLR, 2020

    2020

  9. [17]

    Improved regret analysis for variance-adaptive linear bandits and horizon-free linear mixture mdps

    Kim, Y., Yang, I., and Jun, K.-S. Improved regret analysis for variance-adaptive linear bandits and horizon-free linear mixture mdps. Advances in Neural Information Processing Systems, 35: 0 1060--1072, 2022

    2022

  10. [18]

    and Oh, M.-h

    Lee, J. and Oh, M.-h. Nearly minimax optimal regret for multinomial logistic bandit. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024

    2024

  11. [19]

    and Oh, M.-h

    Lee, J. and Oh, M.-h. Combinatorial reinforcement learning with preference feedback. arXiv preprint arXiv:2502.10158, 2025

    2025 arXiv

  12. [20]

    Improved regret bounds of (multinomial) logistic bandits via regret-to-confidence-set conversion

    Lee, J., Yun, S.-Y., and Jun, K.-S. Improved regret bounds of (multinomial) logistic bandits via regret-to-confidence-set conversion. In International Conference on Artificial Intelligence and Statistics, pp.\ 4474--4482. PMLR, 2024 a

    2024

  13. [21]

    A unified confidence sequence for generalized linear models, with applications to bandits

    Lee, J., Yun, S.-Y., and Jun, K.-S. A unified confidence sequence for generalized linear models, with applications to bandits. arXiv preprint arXiv:2407.13977, 2024 b

    2024 arXiv

  14. [22]

    Provably efficient reinforcement learning with multinomial logit function approximation

    Li, L.-F., Zhang, Y.-J., Zhao, P., and Zhou, Z.-H. Provably efficient reinforcement learning with multinomial logit function approximation. arXiv preprint arXiv:2405.17061, 2024

    2024 arXiv

  15. [23]

    Modelling the choice of residential location

    McFadden, D. Modelling the choice of residential location. 1977

    1977

  16. [24]

    and Iyengar, G

    Oh, M.-h. and Iyengar, G. Thompson sampling for multinomial logit contextual bandits. Advances in Neural Information Processing Systems, 32, 2019

    2019

  17. [25]

    and Iyengar, G

    Oh, M.-h. and Iyengar, G. Multinomial logit contextual bandits: Provable optimality and practicality. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pp.\ 9205--9213, 2021

    2021

  18. [26]

    Multinomial logit bandit with linear utility functions

    Ou, M., Li, N., Zhu, S., and Jin, R. Multinomial logit bandit with linear utility functions. In Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, IJCAI-18 , pp.\ 2602--2608. International Joint Conferences on Artificial Intelligence O...

    2018

  19. [27]

    Infinite-horizon reinforcement learning with multinomial logistic function approximation

    Park, J., Kwon, J., and Lee, D. Infinite-horizon reinforcement learning with multinomial logistic function approximation. arXiv preprint arXiv:2406.13633, 2024

    2024 arXiv

  20. [28]

    and Goyal, V

    Perivier, N. and Goyal, V. Dynamic pricing and assortment under a contextual mnl demand. Advances in Neural Information Processing Systems, 35: 0 3461--3474, 2022

    2022

  21. [29]

    M., and Shmoys, D

    Rusmevichientong, P., Shen, Z.-J. M., and Shmoys, D. B. Dynamic assortment optimization with a multinomial logit choice model and capacity constraint. Operations research, 58 0 (6): 0 1666--1680, 2010

    2010

  22. [30]

    and Zeevi, A

    Saur \'e , D. and Zeevi, A. Optimal dynamic assortment planning with demand learning. Manufacturing & Service Operations Management, 15 0 (3): 0 387--404, 2013

    2013

  23. [31]

    Generalized linear bandits with limited adaptivity

    Sawarni, A., Das, N., Barman, S., and Sinha, G. Generalized linear bandits with limited adaptivity. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. URL https://openreview.net/forum?id=FTPDBQuT4G

    2024

  24. [32]

    Composite convex minimization involving self-concordant-like cost functions

    Tran-Dinh, Q., Li, Y.-H., and Cevher, V. Composite convex minimization involving self-concordant-like cost functions. In Modelling, Computation and Optimization in Information Systems and Management Sciences: Proceedings of the 3rd International Conference on Modelling, Comput...

    2015

  25. [33]

    Etude critique de la notion de collectif, volume 3

    Ville, J. Etude critique de la notion de collectif, volume 3. Gauthier-Villars Paris, 1939

    1939

  26. [34]

    and Sugiyama, M

    Zhang, Y.-J. and Sugiyama, M. Online (multinomial) logistic bandit: Improved regret and constant computation cost. Advances in Neural Information Processing Systems, 36, 2024

    2024

  27. [35]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.