Optimal Contextual Pricing and Extensions

Liu, Allen; Leme, Renato Paes; Schneider, Jon

doi:10.1137/1.9781611976465.66

Cited by 11 publications

(24 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Improving the dependence on is an interesting open direction. After a sequence of papers, this dependence is now optimized when all agents are fully rational [11,26,27,33].…”

Section: Discussionmentioning

confidence: 99%

“…With respect to stochastic noise models, [11] extend their results to a stochastic noise model which we also discuss in Section 2.2. 1 A concurrent work by Liu, Paes Leme, and Schneider [26] also extends results for the absolute loss to a stochastic noise model where the feedback is flipped. However, all of the aforementioned works, both the statistical and the binary search approaches, do not extend to adversarial noise models such as the one we consider.…”

mentioning

confidence: 81%

“…For example, in feature-based dynamic pricing [11,26,27,33], say, of Airbnb apartments, each context x describes a particular apartment with components, or features, providing the apartment's location, cleanliness, and so on. The true value is the price an incoming customer or agent is willing to pay, which is assumed to be a linear function (defined by ★ ) of x .…”

Section: Introductionmentioning

confidence: 99%

“…This line of work was introduced by Cohen, Lobel, and Paes Leme [11] who studied contextual pricing with adversarial contexts. Their results were improved for the -ball loss [27] and the pricing loss [26,33]. Mao, Paes Leme, and Schneider [30] studied a variant of the standard contextual pricing problem, where the buyers' utilities are Lipschitz, rather than linear in the contexts.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Contextual search in the presence of irrational agents

Krishnamurthy

Lykouris

Podimata

et al. 2021

Proceedings of the 53rd Annual ACM SIGACT Symposium on Theory of Computing

View full text Add to dashboard Cite

We study contextual search, a generalization of binary search in higher dimensions, which captures settings such as feature-based dynamic pricing. Standard game-theoretic formulations of this problem assume that agents act in accordance with a specific behavioral model. In practice, some agents may not subscribe to the dominant behavioral model or may act in ways that seem to be arbitrarily irrational. Existing algorithms heavily depend on the behavioral model being (approximately) accurate for all agents and have poor performance even with a few arbitrarily irrational agents.We initiate the study of contextual search when some of the agents can behave in ways inconsistent with the underlying behavioral model. In particular, we provide two algorithms, one based on multidimensional binary search methods and one based on gradient descent. Our techniques draw inspiration from learning theory, game theory, high-dimensional geometry, and convex analysis.

show abstract

“…Improving the dependence on is an interesting open direction. After a sequence of papers, this dependence is now optimized when all agents are fully rational [11,26,27,33].…”

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 81%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Contextual search in the presence of irrational agents

Krishnamurthy

Lykouris

Podimata

et al. 2021

Proceedings of the 53rd Annual ACM SIGACT Symposium on Theory of Computing

View full text Add to dashboard Cite

show abstract

“…However, these methods are not applicable when there are thousands of highly differentiated products with no experience in selling them. This motivates the idea of "contextual pricing" (Cohen et al, 2020;Mao et al, 2018;Javanmard and Nazerzadeh, 2019;Liu et al, 2021), where each sale session is described by a context that also affects the valuation and pricing.…”

Section: Introductionmentioning

confidence: 99%

Towards Agnostic Feature-based Dynamic Pricing: Linear Policies vs Linear Valuation with Unknown Noise

Xu¹,

Wang²

2022

Preprint

View full text Add to dashboard Cite

In feature-based dynamic pricing, a seller sets appropriate prices for a sequence of products (described by feature vectors) on the fly by learning from the binary outcomes of previous sales sessions ("Sold" if valuation ≥ price, and "Not Sold" otherwise). Existing works either assume noiseless linear valuation or preciselyknown noise distribution, which limits the applicability of those algorithms in practice when these assumptions are hard to verify. In this work, we study two more agnostic models: (a) a "linear policy" problem where we aim at competing with the best linear pricing policy while making no assumptions on the data, and (b) a "linear noisy valuation" problem where the random valuation is linear plus an unknown and assumption-free noise. For the former model, we show a Θ(d) minimax regret up to logarithmic factors. For the latter model, we present an algorithm that achieves an Õ(T 3 4 ) regret, and improve the best-known lower bound from Ω(T 3 5 ) to Ω(T 2 3 ). These results demonstrate that no-regret learning is possible for feature-based dynamic pricing under weak assumptions, but also reveal a disappointing fact that the seemingly richer pricing feedback is not significantly more useful than the bandit-feedback in regret reduction.Contextual pricing. For t = 1, 2, ..., T :revealed that describes a sales session (product, customer and context). 2. The customer valuates the product as yt using xt. 3. The seller proposes a price vt > 0 concurrently (according to xt and historical sales records).4. The transaction is successful if vt ≤ yt, i.e., the seller gets a reward (payment) of rt = vt • 1(vt ≤ yt).

show abstract

Policy Optimization Using Semiparametric Models for Dynamic Pricing

Fan

Guo

2022

Journal of the American Statistical Association

View full text Add to dashboard Cite

Optimal Contextual Pricing and Extensions

Cited by 11 publications

References 0 publications

Contextual search in the presence of irrational agents

Contextual search in the presence of irrational agents

Towards Agnostic Feature-based Dynamic Pricing: Linear Policies vs Linear Valuation with Unknown Noise

Policy Optimization Using Semiparametric Models for Dynamic Pricing

Contact Info

Product

Resources

About