Pith. sign in

REVIEW 3 major objections 4 minor 29 references

Hybrid Advertising in the Sponsored Search

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Letting an ad slot show either an independent store or a store-brand bundle, with the bundle count capped at C, raises platform revenue beyond either pure format.

desk verdict A practical, well-executed extension of neural auction design to hybrid store/bundle formats, but the near-DSIC claim rests on a circular and non-exhaustive regret check, so treat revenue as conditional on near-truthful bidding. read the letter →

arxiv 2507.07711 v1 pith:3BZCLZH5 submitted 2025-07-10 cs.GT

Zhen Zhang , Weian Li , Yuhan Wang , Qi Qi , Kun Huang This is my paper
classification cs.GT MSC 91B2668T07
keywords hybridadvertisingsponsoredsearchauctionmechanismdesignautomatedneuralnetworkarchitectureregretminimizationincentivecompatibilityindividualrationality
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

This paper tries to establish that online ad platforms can raise revenue by mixing two ad formats in one auction: each slot can go to an independent store or to a store-brand bundle, with the number of bundles capped at C. The claim is that a learned mechanism, HRegNet, can allocate slots and set payments for both formats while keeping each advertiser's incentive to misreport negligible and keeping truthful participation non-negative. On synthetic distributions and on three days of real e-commerce auction logs, the paper reports that this hybrid mechanism beats store-only, bundle-only, and welfare-maximizing baselines in revenue. If right, the contribution is a practical template for ad systems that want the reach of store ads and the two-sided revenue of bundle ads in a single ranked list.

What carries the argument

The load-bearing object is HRegNet's allocation-and-payment construction. Two subnetworks score independent stores and bundles separately; row-wise and column-wise softmax operations on the raw scores, followed by an elementwise minimum, produce a doubly stochastic allocation matrix that respects the one-slot-per-advertiser and one-advertiser-per-slot constraints. A C-softmax normalization caps the total probability mass assigned to bundles at C. Payments are computed by scaling each advertiser's expected allocated value by a sigmoid parameter in [0,1], which the paper argues keeps payments below bids and hence enforces individual rationality. Training minimizes negative expected revenue plus an augmented Lagrangian penalty on empirical ex-post regret, with optimal misreports found by gradient ascent.

What would settle it

Run a one-slot hybrid auction with one store having $\alpha_i = 1.5$, one bundle containing it, and network probabilities of 0.7 for the independent store and 0.4 for the bundle; the store's expected allocated value is $1.45$, so a sigmoid payment scalar near 1 produces a payment above the store's value, directly contradicting the paper's IR argument if such a trained run exists.

Watch

Extended reading notes

Core claim

The central discovery claimed is that hybrid allocation is not a compromise but a revenue improvement: allowing each slot to be either an independent store or a bundle lets the platform collect from brands as well as stores while still serving users who prefer store-centric results. The paper formalizes this as a constrained mechanism-design problem and solves it with a neural network that takes store and brand value samples as input and outputs an allocation probability matrix and payments. The resulting mechanisms are reported to have average empirical ex-post regret below 0.001 on synthetic tests and a relative regret ratio below 0.065 on real data, with empirical revenue higher than every tested baseline across the reported settings and value distributions. The authors present this as a demonstrated property of the trained mechanism, not as a closed-form theorem.

Load-bearing premise

The individual-rationality guarantee rests on the assumption that each participant's expected allocated value, scaled by a sigmoid parameter in [0,1], can never exceed their bid; this can fail when a store with quality factor above 1 wins both an independent slot and a bundle slot in the same auction.

Editorial extensions

If this is right

  • Platforms can run one auction for both ad formats rather than choosing between store-centric and bundle-centric displays; the paper's experiments indicate the mixed format alone outperforms both pure formats in revenue.
  • The cap C acts as a platform dial: reported revenue rises as more bundles are allowed, so a platform can trade the ad-format mix against user experience by choosing C.
  • The learned mechanism is fast enough for online use: the paper reports roughly 1 to 3 milliseconds per auction sample at inference after 2 to 4 hours of offline training.
  • The revenue gain over a welfare-maximizing baseline comes with lower social welfare in the reported numbers, so a platform that weights advertiser welfare faces a real tradeoff.
  • The design pattern extends, per the paper, to slots displaying independent brands, exclusive displays, or bundles with more than two components with only minor modifications.

Reading between the lines

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

  • The paper leaves implicit that the two-subnetwork pattern is a general recipe for mixed-format auctions: any two display classes with different click-through rates could be handled the same way, not just stores and bundles.
  • The claimed individual-rationality guarantee is only as solid as the premise that expected allocated value stays at or below 1; a store with quality factor above 1 winning both an independent slot and a bundle slot violates that premise, so high-alpha stress tests are the natural next check.
  • A concrete extension would freeze a trained payment network and sweep alpha values above 1 to see whether any sigmoid scalar saturates near 1 while the expected allocated value exceeds 1; that is the point where an IR violation would show up empirically.
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 / 4 minor

Summary. The paper proposes a new sponsored-search advertising model, "hybrid advertising," in which each ad slot can be allocated either to an independent store or to a store-brand bundle, with at most C winning bundles. To compute a revenue-maximizing mechanism in this model, the authors introduce HRegNet, a neural-network architecture whose allocation network enforces the per-slot, per-bidder, and bundle-cap constraints, and whose payment network scales expected allocated values by sigmoid parameters to enforce individual rationality. The training objective is an augmented Lagrangian of negative revenue plus per-advertiser empirical regret penalties (Section 3.2). Experiments on synthetic settings (Tables 1–2) and on real e-commerce log data (Table 3) report that HRegNet achieves significantly higher revenue than RegretNet, JRegNet, VCG, and a revised JRegNet baseline, while reporting regret below 0.001 in synthetic tests and a regret/utility ratio below 0.065 in real-data tests. The paper's central claims are that HRegNet generates near-DSIC and IR mechanisms with superior revenue.

Significance. If the empirical claims withstand scrutiny, the paper makes a useful applied contribution: it defines a practical auction format that is already close to industry practice, encodes nontrivial allocation constraints in a differentiable way, and provides real-data evidence that the hybrid format can outperform both store-only and bundle-only designs. The inclusion of multiple baselines, especially the revised JRegNet, is a strength, and the real-world evaluation on e-commerce log data adds credibility. However, the paper does not provide code or an independent verification protocol, and its theoretical guarantees are only as strong as the regret estimates used during training and testing. The claimed incentive guarantees therefore need additional support before the revenue-improvement headline can be fully relied upon.

major comments (3)
  1. [Section 3.2 / Section 4.1, Eqs. (2)–(4), Algorithm 1, Tables 1–3] The only evidence for the central "near-DSIC" claim is the empirical regret computed with the same procedure that is minimized during training: for each test profile, 100 misreports are initialized and 200 gradient-ascent steps are run, and the maximum over those restarts is reported. For the nonconcave misreport-utility functions induced by a neural allocation and payment rule, a finite number of gradient-ascent restarts yields a lower bound on the true worst-case regret, not an upper bound. A reported regret below 0.001 therefore does not certify near-DSIC, and the revenue numbers in Tables 1–3 are computed under truthful reports that may not be incentive-compatible. I request an independent verification at least in the smallest setting (Setting A), such as an exhaustive grid or exact optimization over the single-dimensional misreport space, or a provable upper bound derived from the network architecture; the same metric used as the training penalty is not an independent certificate.
  2. [Section 3.1, Lemma 1 and the construction of S3] The statement that the matrix S3 = min(column-softmax(S1), row-softmax(S2)) is "doubly stochastic" is not correct for the rectangular matrices used here. The column-wise softmax gives column sums of 1, and the row-wise softmax gives row sums of 1 including the non-allocation column; after taking the elementwise minimum and deleting that column, the row and column sums are at most 1 but are not equal to 1 in general. For example, when the number of candidate rows R+m differs from K, equality of both total sums is impossible. The feasibility constraints (a) and (b) only require inequalities, so the allocation construction can still be repaired, but the lemma as stated and its citation to [8] do not establish the claimed property. This needs a corrected statement and proof for a doubly sub-stochastic matrix.
  3. [Section 3.1, payment network and IR discussion] The sentence "Since we focus on DSIC mechanisms, and \tilde{p}_{i\cdot} \in [0,1] and \tilde{p}_{\cdot j} \in [0,1], the payments cannot exceed the values (i.e., bids), which implies that IR condition is satisfied" is false as written. Because the independent-store allocation term includes the quality factor \alpha_i, and because a store can win both an independent slot and bundle slots (Footnote 3), \sum_k a_{i\cdot k} q_{i\cdot k} can exceed b_i; payments can then exceed the bid. The correct IR argument is that, under truthful reporting, payment equals \tilde{p}_{i\cdot} v_i g_i, so truthful utility is (1-\tilde{p}_{i\cdot}) v_i g_i \ge 0. The proof should be corrected, and the model should clarify whether a store may appear in multiple winning bundles simultaneously, since this affects both the IR argument and the interpretation of the allocation constraints.
minor comments (4)
  1. [Section 2.2, definition of ex-post regret] The displayed definition of rgti·(v) includes an expectation over v ∼ F while also depending on a fixed value profile v; this appears to conflate ex-post regret with expected regret. The empirical version in Eq. (2) is clearer, and the definition should be rewritten to match it.
  2. [Section 2.1 and Section 3.1, notation] The notation for bids and values is inconsistent: the mechanism is defined with bids b, but payment rules are written as p : V → R≥0, and the utility formulas mix b and v. Aligning the notation would remove avoidable confusion about which arguments are truthful reports.
  3. [Table 3] The DSIC metric ru is reported only as a uniform bound "<0.065" for every method and every C value, which makes the column uninformative for comparing methods. The actual ratios, or at least their standard deviations, should be reported.
  4. [Section 4.1] The test-time regret evaluation uses 100 restarts and 200 gradient steps, but no details are given about the step size, the initialization distribution, or the stopping rule for the gradient-ascent misreport search; these details are necessary for other researchers to reproduce the reported regret values.

Circularity Check

1 steps flagged · score 4.0 of 10

Near-DSIC support re-measures the training regret objective; the headline revenue improvement is an independent empirical comparison, so circularity is partial rather than total.

  1. fitted input called prediction [Section 3.2 (Eq. 2 and Eq. 4), Section 4 Evaluation metric, Section 4.1 testing procedure, Tables 1-3]
    "The average empirical ex-post regret: 1/(n+m) (Σ_{i=1}^m crgt_i· + Σ_{j=1}^n crgt_·j)."

    The regret reported as the near-DSIC guarantee is exactly the empirical ex-post regret crgt defined in Eq. (2), which the augmented-Lagrangian objective C_ρ(w;λ) in Section 3.2 penalizes; the gradient in Eq. (4) contains ρ·frgt(w)·g terms, so training drives this quantity toward zero. At test time, Section 4.1 recomputes the same quantity using the same gradient-ascent heuristic (100 misreport restarts, 200 iterations) used in training. If the heuristic misses the true optimal misreport for a nonconcave neural utility, the measured regret is a lower bound, not an independent certificate of DSIC. The 'near-DSIC' claim therefore restates the training objective rather than testing it against an independent, exact or exhaustive benchmark.

full rationale

The headline claim — that HRegNet's hybrid mechanisms improve revenue over RegretNet, JRegNet, revised JRegNet, and VCG — is an empirical comparison on held-out synthetic and real data, not a derivation from a fitted parameter, so it is not circular. The main circularity concern is the near-DSIC validation: the average empirical ex-post regret used in Tables 1-3 is the same crgt objective that HRegNet minimizes during training (Eqs. 2-4), and it is evaluated with the same finite-restart gradient ascent used to find misreports during training; this makes the low regret numbers a partial re-measurement of the optimization target rather than an independent guarantee. Separately, Section 3.1's claim that p~∈[0,1] makes 'payments cannot exceed the values' is literally false when α_i>1 and a store wins both an independent slot and a bundle slot, since the expected allocated value can exceed 1; however, IR still holds by construction because p_i = p~_i·v_i·g_i gives truthful utility (1-p~_i)v_i·g_i ≥ 0, so this is a proof gap, not circularity. The self-citation of JRegNet [28] is disclosed and used mainly as a baseline and motivation; the paper's own experiments also evaluate joint advertising, so the citation is not load-bearing. Overall, the central revenue result has independent content; only the DSIC support is partially self-referential, giving a score of 4.

Assumptions & free parameters 3 free parameters · 6 assumptions · 1 invented entities

The mechanism is entirely learned: allocation and payments are neural network outputs with no closed-form characterization, so the fitted network weights and payment scalars are the central free parameters. The structural constraints come from softmax normalizations and the design choice C. The paper contributes a new model and an empirical method, not a derivation with few free parameters.

free parameters (3)
  • Network parameters w = Learned; no values reported
    Allocation and payment functions are outputs of a trained neural network; the reported revenue and regret results depend on these fitted weights.
  • Payment scaling parameters p~_i and p~_j = Learned sigmoid outputs in [0,1]; no values reported
    Core of the payment rule: each payment is a fitted fraction of the advertiser's expected allocated value. IR is claimed from this parameterization.
  • Maximum winning bundles C = Swept as C = 1 to 5 in experiments
    A design choice that changes the feasible set; revenue rankings depend on C.
assumptions (6)
  • domain assumption Advertiser values are independently sampled from known distributions F_i and F_j.
    Section 2.1: needed to define expected revenue and to train on samples; the paper does not handle correlated or unknown value distributions.
  • domain assumption Click-through rates factor as alpha_i * theta_k for independent stores and theta_k for bundles, with alpha_i > 0 and exogenous position CTRs theta_k.
    Section 2.1: this separability underlies the allocation objective, the payment formula, and the comparison across ad formats.
  • domain assumption A store may win an independent slot and a bundle slot in the same auction (roles are not mutually exclusive).
    Footnote 3: this assumption, combined with alpha_i > 1, breaks the claimed IR guarantee.
  • standard math The phi_BS transformation yields a doubly stochastic matrix and, combined with the C-normalized matrix H', yields an allocation satisfying constraints (a)-(c).
    Lemma 1 is cited from [8]; it is used to enforce row, column, and bundle-count feasibility.
  • domain assumption Gradient ascent over misreports converges to the true optimal misreport, and the empirical regret approximates worst-case regret.
    Sections 3.2 and 4.1: near-DSIC is measured entirely through this procedure.
  • ad hoc to paper A fixed-size neural network can represent or approximate the revenue-optimal hybrid auction.
    No representability theorem is given; optimality is inferred from training success on finite samples.
invented entities (1)
  • Hybrid advertising auction format with a C-bundle cap
    purpose: Allows each ad slot to be allocated to either an independent store or a store-brand bundle, with at most C bundle winners.
    Proposed in this paper; its value is supported only by in-paper experiments, with no external benchmark or formal characterization.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hybrid Advertising in the Sponsored Search." pith.science (2026). https://pith.science/paper/3BZCLZH5

@misc{pith2026250707711,
  author       = {Pith},
  title        = {Pith review of: Hybrid Advertising in the Sponsored Search},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3BZCLZH5}},
  note         = {Machine review of arXiv:2507.07711}
}
read the original abstract

Online advertisements are a primary revenue source for e-commerce platforms. Traditional advertising models are store-centric, selecting winning stores through auction mechanisms. Recently, a new approach known as joint advertising has emerged, which presents sponsored bundles combining one store and one brand in ad slots. Unlike traditional models, joint advertising allows platforms to collect payments from both brands and stores. However, each of these two advertising models appeals to distinct user groups, leading to low click-through rates when users encounter an undesirable advertising model. To address this limitation and enhance generality, we propose a novel advertising model called ''Hybrid Advertising''. In this model, each ad slot can be allocated to either an independent store or a bundle. To find the optimal auction mechanisms in hybrid advertising, while ensuring nearly dominant strategy incentive compatibility and individual rationality, we introduce the Hybrid Regret Network (HRegNet), a neural network architecture designed for this purpose. Extensive experiments on both synthetic and real-world data demonstrate that the mechanisms generated by HRegNet significantly improve platform revenue compared to established baseline methods.

Figures

Figures reproduced from arXiv: 2507.07711 by the authors.

Figure 1
Figure 1. Traditional Ad Model, Joint Ad Model and Hybrid Ad Model. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The architecture of HRegNet is designed for a scenario with [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

29 extracted references · 26 canonical work pages

  1. [8]

    Optimal auctions through deep learning

    Paul Dütting, Zhe Feng, Harikrishna Narasimhan, David Parkes, and Sai Srivatsa Ravindranath. Optimal auctions through deep learning. In International Conference on Machine Learning, pages 1706–1715. PMLR, 2019

    work page 2019

  2. [1]

    Selling joint ads: A regret minimization perspective

    Gagan Aggarwal, Ashwinkumar Badanidiyuru, Paul Dütting, and Federico Fusco. Selling joint ads: A regret minimization perspective. arXiv preprint arXiv:2409.07819, 2024

    arXiv 2024

  3. [2]

    Reducing mechanism design to algorithm design via machine learning

    Maria-Florina Balcan, Avrim Blum, Jason D Hartline, and Yishay Mansour. Reducing mechanism design to algorithm design via machine learning. Journal of Computer and System Sciences, 74(8):1245–1270, 2008

    work page 2008

  4. [3]

    On the efficiency of equilibria in generalized second price auctions

    Ioannis Caragiannis, Christos Kaklamanis, Panagiotis Kanellopoulos, and Maria Kyropoulou. On the efficiency of equilibria in generalized second price auctions. In Proceedings of the 12th ACM conference on Electronic commerce, pages 81–90, 2011

    work page 2011

  5. [4]

    Multipart pricing of public goods

    Edward H Clarke. Multipart pricing of public goods. Public choice, pages 17–33, 1971

    work page 1971

  6. [5]

    Complexity of Mechanism Design

    Vincent Conitzer and Tuomas Sandholm. Complexity of mechanism design. arXiv preprint cs/0205075, 2002

    work page Pith review arXiv 2002

  7. [6]

    Self-interested automated mechanism design and implications for optimal combinatorial auctions

    Vincent Conitzer and Tuomas Sandholm. Self-interested automated mechanism design and implications for optimal combinatorial auctions. In Proceedings of the 5th ACM Conference on Electronic Commerce , pages 132–141, 2004

    work page 2004

  8. [7]

    A context- integrated transformer-based neural network for auction design

    Zhijian Duan, Jingwu Tang, Yutong Yin, Zhe Feng, Xiang Yan, Manzil Zaheer, and Xiaotie Deng. A context- integrated transformer-based neural network for auction design. InInternational Conference on Machine Learning, pages 5609–5626. PMLR, 2022

    work page 2022

Show all 29 references
  1. [9]

    Optimal auctions through deep learning: Advances in differentiable economics

    Paul Dütting, Zhe Feng, Harikrishna Narasimhan, David C Parkes, and Sai Srivatsa Ravindranath. Optimal auctions through deep learning: Advances in differentiable economics. Journal of the ACM, 71(1):1–53, 2024

    2024

  2. [10]

    Payment rules through discriminant-based classifiers, 2015

    Paul Dütting, Felix Fischer, Pichayut Jirapinyo, John K Lai, Benjamin Lubin, and David C Parkes. Payment rules through discriminant-based classifiers, 2015

    2015

  3. [11]

    Internet advertising and the generalized second- price auction: Selling billions of dollars worth of keywords

    Benjamin Edelman, Michael Ostrovsky, and Michael Schwarz. Internet advertising and the generalized second- price auction: Selling billions of dollars worth of keywords. American economic review, 97(1):242–259, 2007

    2007

  4. [12]

    Deep learning for revenue-optimal auctions with budgets

    Zhe Feng, Harikrishna Narasimhan, and David C Parkes. Deep learning for revenue-optimal auctions with budgets. In Proceedings of the 17th international conference on autonomous agents and multiagent systems, pages 354–362, 2018

    2018

  5. [13]

    Bayes-nash equilibria of the generalized second price auction

    Renato D Gomes and Kane S Sweeney. Bayes-nash equilibria of the generalized second price auction. In Proceedings of the 10th ACM conference on Electronic Commerce, pages 107–108, 2009

    2009

  6. [14]

    Incentives in teams

    Theodore Groves. Incentives in teams. Econometrica: Journal of the Econometric Society, pages 617–631, 1973

    1973

  7. [15]

    Optimal-er auctions through attention

    Dmitry Ivanov, Iskander Safiulin, Igor Filippov, and Ksenia Balabaeva. Optimal-er auctions through attention. Advances in Neural Information Processing Systems, 35:34734–34747, 2022

    2022

  8. [16]

    Proportionnet: Balancing fairness and revenue for auction design with deep learning

    Kevin Kuo, Anthony Ostuni, Elizabeth Horishny, Michael J Curry, Samuel Dooley, Ping-yeh Chiang, Tom Goldstein, and John P Dickerson. Proportionnet: Balancing fairness and revenue for auction design with deep learning. arXiv preprint arXiv:2010.06398, 2020

    2010 arXiv

  9. [17]

    A kernel-based iterative combinatorial auction

    Sébastien Lahaie. A kernel-based iterative combinatorial auction. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 25, pages 695–700, 2011

    2011

  10. [18]

    Revenue analysis of a family of ranking rules for keyword auctions

    Sébastien Lahaie and David M Pennock. Revenue analysis of a family of ranking rules for keyword auctions. In Proceedings of the 8th ACM Conference on Electronic Commerce, pages 50–56, 2007

    2007

  11. [19]

    On revenue in the generalized second price auction

    Brendan Lucier, Renato Paes Leme, and Éva Tardos. On revenue in the generalized second price auction. In Proceedings of the 21st international conference on World Wide Web, pages 361–370, 2012

    2012

  12. [20]

    Joint bidding in ad auctions

    Yuchao Ma, Weian Li, Wanzhi Zhang, Yahui Lei, Zhicheng Zhang, Qi Qi, Qiang Liu, and Xingxing Wang. Joint bidding in ad auctions. In Annual Conference on Theory and Applications of Models of Computation , pages 344–354. Springer, 2024

    2024

  13. [21]

    Optimal auction design

    Roger B Myerson. Optimal auction design. Mathematics of operations research, 6(1):58–73, 1981

    1981

  14. [22]

    Preferencenet: Encoding human preferences in auction design with deep learning

    Neehar Peri, Michael Curry, Samuel Dooley, and John Dickerson. Preferencenet: Encoding human preferences in auction design with deep learning. Advances in Neural Information Processing Systems, 34:17532–17542, 2021

    2021

  15. [23]

    A permutation-equivariant neural network architecture for auction design

    Jad Rahme, Samy Jelassi, Joan Bruna, and S Matthew Weinberg. A permutation-equivariant neural network architecture for auction design. In Proceedings of the AAAI conference on artificial intelligence, volume 35, pages 5664–5672, 2021. 13

    2021

  16. [24]

    Automated mechanism design: A new application area for search algorithms

    Tuomas Sandholm. Automated mechanism design: A new application area for search algorithms. In International Conference on Principles and Practice of Constraint Programming, pages 19–36. Springer, 2003

    2003

  17. [25]

    Automated design of revenue-maximizing combinatorial auctions

    Tuomas Sandholm and Anton Likhodedov. Automated design of revenue-maximizing combinatorial auctions. Operations Research, 63(5):1000–1025, 2015

    2015

  18. [26]

    Revenue optimization in the generalized second-price auction

    David RM Thompson and Kevin Leyton-Brown. Revenue optimization in the generalized second-price auction. In Proceedings of the fourteenth ACM conference on Electronic commerce, pages 837–852, 2013

    2013

  19. [27]

    Counterspeculation, auctions, and competitive sealed tenders.The Journal of finance, 16(1):8–37, 1961

    William Vickrey. Counterspeculation, auctions, and competitive sealed tenders.The Journal of finance, 16(1):8–37, 1961

    1961

  20. [28]

    Joint auction in the online advertising market

    Zhen Zhang, Weian Li, Yahui Lei, Bingzhe Wang, Zhicheng Zhang, Qi Qi, Qiang Liu, and Xingxing Wang. Joint auction in the online advertising market. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 4362–4373, 2024

    2024

  21. [29]

    Deep interest network for click-through rate prediction

    Guorui Zhou, Xiaoqiang Zhu, Chenru Song, Ying Fan, Han Zhu, Xiao Ma, Yanghui Yan, Junqi Jin, Han Li, and Kun Gai. Deep interest network for click-through rate prediction. In Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining, pages...

    2018

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.