A Robust Non-Clairvoyant Dynamic Mechanism for Contextual Auctions

Authors: Yuan Deng, Sébastien Lahaie, Vahab Mirrokni

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper, we present a framework of designing non-clairvoyant dynamic mechanisms that are robust to both the estimation errors on the buyer s distributional information and the buyer s strategic behavior. We then tailor our framework to the setting of contextual auctions to develop a non-clairvoyant mechanism that achieves no-regret against 1/3-approximation of the revenue-optimal clairvoyant dynamic mechanism. The paper focuses on theoretical contributions, including mechanism design, mathematical proofs (Theorem 3.1, Lemma 3.1-3.6), and analysis of regret bounds (O(T^5/6)), without reporting any empirical studies, datasets, or experimental results.
Researcher Affiliation Collaboration Yuan Deng Duke University Durham, NC ericdy@cs.duke.edu; Sébastien Lahaie Google Research New York City, NY slahaie@google.com; Vahab Mirrokni Google Research New York City, NY mirrokni@google.com
Pseudocode Yes Figure 1: Robust Non-clairvoyant Dynamic Contextual Auction Policy. This figure presents a structured description of the 'Learning Policy' and 'Dynamic Mechanism Policy' in a step-by-step format, which functions as an algorithm block.
Open Source Code No The paper does not contain any explicit statements or links indicating that open-source code for the described methodology is provided.
Open Datasets No The paper is theoretical and describes a problem framework (e.g., 'observable feature vector ζt Rd', 'feature vectors are drawn independently from a fixed distribution D') rather than using specific, named datasets or providing access information for any dataset. Therefore, there is no mention of publicly available or open datasets.
Dataset Splits No The paper is theoretical and does not conduct experiments with empirical data. As such, it does not provide details on dataset splits (training, validation, test) needed for reproducibility.
Hardware Specification No The paper is theoretical and does not describe any experimental setup or computational execution. Therefore, no hardware specifications are mentioned.
Software Dependencies No The paper is theoretical and does not mention any specific software dependencies or versions used in its development.
Experiment Setup No The paper is theoretical and does not describe an empirical experimental setup, including hyperparameters or specific training settings.