arXiv AI
Randomly Wrong Signals: Bayesian Auction Design with ML Predictions
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Sep 4
Data Market Design through Deep Learning
The paper tackles the data market design problem, which seeks signaling schemes that maximize revenue for an information seller. It applies deep learning to learn these schemes, addressing both obedience and incentive constraints, and demonstrates that the framework can replicate known theoretical solutions, extend to more complex scenarios, and suggest new optimal designs. The study builds on prior auction‑design work and introduces a novel approach for revenue‑optimal data markets.
arXiv Machine Learning
Oct 23
Fast Iterative Combinatorial Auctions via Deep Reinforcement Learning
FastCombo is a deep reinforcement learning method for iterative combinatorial auctions that learns price policies offline and applies them online with a single neural‑network forward pass. By formulating the problem as a partially observable Markov decision process, it assigns credit to queries that enable future decisions, thereby reducing the expected number of auction rounds. Experiments show that FastCombo achieves competitive clearing while cutting online price‑selection time by roughly 300–900 times compared to a standard Bayesian baseline.
arXiv Machine Learning
Aug 18
Learning to Price with Persuasion
arXiv:2608. 16699v1 Announce Type: cross Abstract: Motivated by modern marketplaces, where the platform or the seller routinely gathers detailed user profiles, we study a novel learning theoretic model that simultaneously involves information and mechanism design.
arXiv Machine Learning
Oct 19
Source-Aware Dynamic Pricing with Strategic Buyers
arXiv:2610.22516v1 Announce Type: cross Abstract: In modern commerce, firms and platforms typically use customer features from different sources to personalize prices: some features are verified usin...
arXiv Machine Learning
Jul 22
JD-BP: A Joint-Decision Generative Framework for Auto-Bidding and Pricing
arXiv:2604. 05845v2 Announce Type: replace-cross Abstract: Auto-bidding services optimize real-time bidding strategies for advertisers under key performance indicator (KPI) constraints such as target return on investment and budget.
arXiv Machine Learning
Oct 23
LLM-Auction: Generative Auction towards LLM-Native Advertising
The paper introduces LLM‑Auction, a learning‑based generative auction mechanism that integrates ad allocation with large language model (LLM) generation. It redefines the auction object from discrete ad slots to distributions over LLM outputs, aligning allocation with advertiser value and user experience while inherently modeling allocation externalities without extra inference cost. The authors prove allocation monotonicity and continuity, show a first‑price payment rule has favorable incentive properties, and demonstrate state‑of‑the‑art allocation efficiency in a simulated LLM‑as‑a‑judge environment.