Abstract:For decades, Simultaneous Ascending Auction (SAA) has been the most widely used mechanism for spectrum auctions, and it has recently gained popularity for allocating 5G licenses in many countries. Despite its relatively simple rules, SAA introduces a complex strategic game with an unknown optimal bidding strategy. Given the high stakes involved, with billions of euros sometimes on the line, developing an efficient bidding strategy is of utmost importance. In this work, we extend our previous method, a Simultaneous Move Monte-Carlo Tree Search (SM-MCTS) based algorithm named $SMS^{\alpha}$ to incomplete information framework. For this purpose, we compare three determinization approaches which allow us to rely on complete information SM-MCTS. This algorithm addresses, in incomplete framework, the four key strategic issues of SAA: the exposure problem, the own price effect, budget constraints, and the eligibility management problem. Through extensive numerical experiments on instances of realistic size with an uncertain framework, we show that $SMS^{\alpha}$ largely outperforms state-of-the-art algorithms by achieving higher expected utility while taking less risks, no matter which determinization method is chosen.

What problem does this paper attempt to address?

This paper attempts to solve the complex strategic problems caused by incomplete information in the Simultaneous Ascending Auction (SAA). Specifically, SAA is a mechanism widely used in spectrum auctions, especially when allocating 5G licenses. Although its rules are relatively simple, SAA introduces a complex strategic game in which the optimal bidding strategy is still unclear. Given the huge amounts involved in auctions (sometimes reaching billions of euros), it is crucial to develop efficient bidding strategies. The main objective of the paper is to solve four key strategic problems in SAA by using Monte Carlo Tree Search (MCTS) and determinization methods: 1. **Exposure Problem**: When bidders bid for a set of complementary licenses, they may pay too high a price due to overly intense competition. 2. **Own Price Effect**: As the price rises, the bidder's utility decreases, so the bidder may choose not to bid on certain licenses to keep the price low. 3. **Budget Constraints**: Bidders have limited funds throughout the auction process, and improper fund allocation may lead to sub - optimal results. 4. **Eligibility Management Problem**: The activity rules require bidders to maintain a minimum level of bidding activity, otherwise they will be penalized, which affects the range of items that bidders can bid on. To address these challenges, the authors propose three determinization methods and combine them with the Simultaneous Move Monte Carlo Tree Search (SM - MCTS) algorithm to handle the situation of incomplete information. These methods aim to increase the expected utility and reduce risks, and no matter which determinization method is chosen, it can be significantly superior to existing algorithms. ### Key Contributions - **Modeling**: Model the auction as a multi - player simultaneous - move game with incomplete information (SAA - inc) without making specific assumptions about the value function. - **Strategy**: Propose three efficient bidding strategies while solving the four main strategic problems in SAA - inc. - **Determinization Methods**: Introduce determinization methods based on expectation, separation tree, and single tree, expanding the SM - MCTS algorithm that was previously only applicable to the complete - information case. - **Experimental Verification**: Through a large number of numerical experiments, prove that the proposed determinization methods are significantly superior to existing algorithms in budget - and eligibility - constrained environments. ### Summary This paper successfully solves the complex strategic problems brought by incomplete information in SAA by introducing new determinization methods and an improved MCTS algorithm, providing more efficient bidding strategies for spectrum auctions.

Bidding efficiently in Simultaneous Ascending Auctions with incomplete information using Monte Carlo Tree Search and determinization

Bidding efficiently in Simultaneous Ascending Auctions with budget and eligibility constraints using Simultaneous Move Monte Carlo Tree Search

TAMES: A Truthful Auction Mechanism for Heterogeneous Spectrum Allocation

Multi-seller Combinatorial Spectrum Auction with Reserve Prices

Many-To-Many Matching for Combinatorial Spectrum Trading

Bidding Privacy Preservation for Dynamic Matching Based Spectrum Trading

TAHES: A Truthful Double Auction Mechanism for Heterogeneous Spectrums

Stable Combinatorial Spectrum Matching

Bid Privacy Preservation in Matching-Based Multiradio Multichannel Spectrum Trading

MastDP: Matching Based Double Auction Mechanism for Spectrum Trading with Differential Privacy

Robust Large-Scale Spectrum Auctions Against False-Name Bids

An Approximately Strategy-Proof Mechanism for Radio Spectrum Allocation.

A Truthful Double Auction Framework for Promoting Femtocell Access.

Secondary Spectrum Auctions for Markets with Communication Constraints

A Prior-Free Revenue Maximizing Auction for Secondary Spectrum Access

Maximizing Social Welfare in Operator-based Cognitive Radio Networks under Spectrum Uncertainty and Sensing Inaccuracy

Semi-Distributed Joint Power and Spectrum Allocation for LAA Based Small Cell Networks

A Truthful QoS-Aware Spectrum Auction with Spatial Reuse for Large-Scale Networks

Dealing with the Untrustworthy Auctioneer in Combinatorial Spectrum Auctions.

Designing Two-Dimensional Spectrum Auctions for Mobile Secondary Users

Revenue-Maximizing and Truthful Online Auctions for Dynamic Spectrum Access