Improved learning rates in multi-unit uniform price auctions

2025-01-17

AI Summary

This paper studies the online bidding learning problem for power generators in day-ahead electricity markets under multi-unit uniform-price auctions, focusing on adversarial environments and information-limited settings. To address auction-specific structural constraints, we propose the first bid-space modeling method tailored to uniform-price auction mechanisms and design a novel “winning-bid revelation” feedback scheme—a compact, hybrid feedback paradigm lying strictly between full-information and bandit feedback. Leveraging structured regret analysis, we develop an adversarial bandit learning algorithm. Under standard bandit feedback, our algorithm achieves an improved regret bound of $ ilde{O}(K^{4/3}T^{2/3})$, surpassing the prior best $ ilde{O}(K^{7/4}T^{3/4})$. With the new winning-bid revelation feedback, we attain a tight regret bound of $ ilde{O}(K^{5/2}sqrt{T})$, establishing the first near-optimal rate for this setting.

Technology Category

Machine Learning: Online Learning & BanditsGame Theory and Economic Paradigms: Adversarial LearningSearch and Optimization: Learning to Search

Application Category

Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactionsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingUser Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

Motivated by the strategic participation of electricity producers in electricity day-ahead market, we study the problem of online learning in repeated multi-unit uniform price auctions focusing on the adversarial opposing bid setting. The main contribution of this paper is the introduction of a new modeling of the bid space. Indeed, we prove that a learning algorithm leveraging the structure of this problem achieves a regret of $ ilde{O}(K^{4/3}T^{2/3})$ under bandit feedback, improving over the bound of $ ilde{O}(K^{7/4}T^{3/4})$ previously obtained in the literature. This improved regret rate is tight up to logarithmic terms. Inspired by electricity reserve markets, we further introduce a different feedback model under which all winning bids are revealed. This feedback interpolates between the full-information and bandit scenarios depending on the auctions' results. We prove that, under this feedback, the algorithm that we propose achieves regret $ ilde{O}(K^{5/2}sqrt{T})$.

Problem

• Multi-item Auctions
• Learning Bidding Strategies
• Information Limited Environments

Innovation

• New Bidding Strategy
• Learning Methodology
• Improved Decision-making in Auctions