Multi-Product Dynamic Pricing in High-Dimensions with Heterogeneous Price Sensitivity

  title={Multi-Product Dynamic Pricing in High-Dimensions with Heterogeneous Price Sensitivity},
  author={Adel Javanmard and Hamid Nazerzadeh and Simeng Shao},
  journal={2020 IEEE International Symposium on Information Theory (ISIT)},
We consider the problem of multi-product dynamic pricing, in a contextual setting, for a seller of differentiated products. In this environment, the customers arrive over time and products are described by high-dimensional feature vectors. Each customer chooses a product according to the widely used Multinomial Logit (MNL) choice model and her utility depends on the product features as well as the prices offered. The seller a-priori does not know the parameters of the choice model but can learn… 
Distribution-free Contextual Dynamic Pricing
This paper establishes the regret upper bound and a matching lower bound of the policy in the perturbed linear bandit framework and proves a sub-linear regret bound in the considered pricing problem.
Dynamic pricing and assortment under a contextual MNL demand
A randomized dynamic pricing policy based on a variant of the Online Newton Step algorithm (ONS) that achieves a O(d √) over T periods is proposed.
Synergy between Customer Segmentation and Personalization
This work considers a personalized pricing problem with revisiting customers, and proposes a heuristic that combines the idea of binary search and feature-based pricing, and shows that it outperforms common benchmarks.
Dynamic Incentive-Aware Learning: Robust Pricing in Contextual Auctions
This work proposes two learning policies that are robust to strategic behavior in repeated contextual second-price auctions and uses the outcomes of the auctions, rather than the submitted bids, to estimate the preferences while controlling the long-term effect of the outcome of each auction on the future reserve prices.
Policy Optimization Using Semiparametric Models for Dynamic Pricing
  • Jianqing Fan, Yongyi Guo, Mengxin Yu
  • Computer Science, Economics
    SSRN Electronic Journal
  • 2021
A dynamic statistical learning and decision-making policy is proposed that combines semiparametric estimation from a generalized linear model with an unknown link and online decision making to minimize regret (maximize revenue) and generalize to the case when the product features are dynamically dependent.
Optimal Non-parametric Learning in Repeated Contextual Auctions with Strategic Buyer
A novel non-parametric learning algorithm is introduced that is horizon-independent and has tight strategic regret upper bound of Θ(T ), and several value-localization techniques of noncontextual repeated auctions are non-trivially generalized to make them effective in the considered contextual non- Parametric learning of the buyer valuation function.
Online Regularization for High-Dimensional Dynamic Pricing Algorithms
The online regularization scheme equips the proposed optimistic online regularized maximum likelihood pricing algorithm with three major advantages: encode market noise knowledge into pricing process optimism; empower online statistical learning with always-validity over all decision points; envelop prediction error process with time-uniform non-asymptotic oracle inequalities.
Bisection-Based Pricing for Repeated Contextual Auctions against Strategic Buyer
A novel deterministic learning algorithm that is based on ideas of the Bisection method and has strategic regret upper bound of O(log T) and the regret guarantee holds for any realization of feature vectors (adversarial upper bound).
Online Regularization towards Always-Valid High-Dimensional Dynamic Pricing
  • ChiHua Wang, Zhanyu Wang, Will Wei Sun, Guang Cheng
  • Mathematics, Computer Science
  • 2020
This paper makes the decisive observation that the always-validity of pricing decisions builds and thrives on the online regularization scheme, and proposes an optimistic online Lasso procedure that resolves dynamic pricing problems at the process level, based on a novel use of non-asymptotic martingale concentration.
Dimension Reduction in Contextual Online Learning via Nonparametric Variable Selection
A variable selection algorithm called BV-LASSO is proposed, which incorporates novel ideas such as binning and voting to apply LASSO to nonparametric settings and may serve as a general recipe to achieve dimension reduction via variable selection in non parametric settings.


Dynamic Pricing in High-Dimensions
A dynamic policy is proposed, called Regularized Maximum Likelihood Pricing (RMLP) that leverages the (sparsity) structure of the high-dimensional model and obtains a logarithmic regret in T, and it is shown that no policy can obtain regret better than RMLP.
Personalized Dynamic Pricing with Machine Learning: High Dimensional Features and Heterogeneous Elasticity
We consider a seller who can dynamically adjust the price of a product at the individual customer level, by utilizing information about customers’ characteristics encoded as a d-dimensional feature
Dynamic Pricing Under a General Parametric Choice Model
A stylized dynamic pricing model in which a monopolist prices a product to a sequence of T customers who independently make purchasing decisions based on the price offered according to a general parametric choice model shows that the regret of the optimal pricing policy in this model is $\Theta(\sqrt T)$.
Perishability of Data: Dynamic Pricing under Varying-Coefficient Models
This work proposes a pricing policy based on projected stochastic gradient descent (PSGD) and characterize its regret in terms of time $T, features dimension $d, and the temporal variability in the model parameters, $\delta_t$.
Feature-based Dynamic Pricing
This work considers the problem faced by a firm that receives highly differentiated products in an online fashion and needs to price them in order to sell them to its customer base, and proposes a modification of the prior algorithm where uncertainty sets are replaced by their Lowner-John ellipsoids.
Dynamic Pricing for Nonperishable Products with Demand Learning
The retailer's problem is formulated as a (Poisson) intensity control problem and the structural properties of an optimal solution are derived, and a simple and efficient approximated solution is suggested.
Dynamic Pricing with Demand Covariates
This work assumes that the firm has access to demand covariates which may be predictive of the demand and proves that GILS achieves an asymptotically optimal regret of order log(T), and shows that the asymPTotic optimality of GILS holds even when the covariates are uninformative.
Dynamic Pricing and Demand Learning with Limited Price Experimentation
In a dynamic pricing problem where the demand function is not known a priori, price experimentation can be used as a demand learning tool. Existing literature usually assumes no constraint on price
Dynamic Pricing with a Prior on Market Response
Computer results demonstrate that decay balancing offers significant revenue gains over recently studied certainty equivalent and greedy heuristics, and establish that changes in inventory and uncertainty in the arrival rate bear appropriate directional impacts on decay balancing prices in contrast to these alternatives.
Pricing under the Generalized Extreme Value Models with Homogeneous Price Sensitivity Parameters
We consider pricing problems when customers choose according to the generalized extreme value (GEV) models and the products have the same price sensitivity parameter. First, we consider the static