Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

Non-Stationary Contextual Pricing with Safety Constraints

Authors: Dheeraj Baby, Jianyu Xu, Yu-Xiang Wang

TMLR 2023 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We present an algorithm Pro DR (Algorithm 1) that attains an optimal O(d3(T 1 3 C 2 3 T 1)) dynamic regret (modulo dependencies in d and log T) for the setting of proper OCO with co-variates under exp-concave losses (see Section 3.1). We construct an algorithm PDRP (Algorithm 2) with a base learner Pro DR, which solves the non-stationary contextual pricing problem with strictly log-concave noise. We define the dynamic regret of contextual pricing as Eq.(3) and show that PDRP achieves a O(d3(T 1 3 C 2 3 T 1)) dynamic regret guarantee (see Section 3.2). We show that any algorithm must incur a dynamic regret of Ω(T 1 3 C 2 3 T 1) in the contextual pricing problem, which says that PDRP is minimax optimal up to d and log T factors (see Section 3.3).
Researcher Affiliation Academia Dheeraj Baby EMAIL Department of Computer Science University of California, Santa Barbara; Jianyu Xu EMAIL Department of Computer Science University of California, Santa Barbara; Yu-Xiang Wang EMAIL Department of Computer Science University of California, Santa Barbara
Pseudocode Yes Algorithm 1 Proper Dynamic Regret minimization (Pro DR); Algorithm 2 Proper Dynamic Regret Pricing (PDRP)
Open Source Code No The paper does not contain any explicit statements about releasing source code or provide links to a code repository.
Open Datasets No The paper describes a linear noisy valuation model for customers and market noises drawn from a known strictly log-concave distribution. It does not mention the use of any specific publicly available datasets nor provides access information for any data used.
Dataset Splits No The paper is theoretical and does not conduct experiments on datasets, therefore no dataset splits are mentioned.
Hardware Specification No The paper is theoretical and does not describe any experiments that would require specific hardware, thus no hardware specifications are provided.
Software Dependencies No The paper is theoretical and does not describe any experimental implementation details that would list software dependencies with specific version numbers.
Experiment Setup No The paper is theoretical and focuses on algorithm design and regret analysis rather than empirical evaluation, so it does not include details on experimental setup or hyperparameters.