Combinatorial Auctions via Machine Learning-based Preference Elicitation
Combinatorial Auctions via Machine Learning-based Preference Elicitation
复制标题
通过基于机器学习的偏好诱导进行组合拍卖
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Sven Seuken
中科院分区：
文献类型：
--
作者：
Gianluca Brero;Benjamin Lubin;Sven Seuken
组合拍卖（CA）用于在具有复杂估价的投标人之间分配多个物品。由于价值空间在项目数量上呈指数级增长，即使在中等规模的环境中，投标人也不可能报告其全部价值函数。先前的工作表明，目前的设计往往无法引出投标人最相关的价值，从而导致效率低下。我们通过引入基于机器学习的启发算法来确定从投标人查询哪些值来解决这个问题。基于这一启发范式，我们设计了一个新的CA机制，我们称之为PVM，支付确定，使投标人的激励与配置效率。我们验证PVM实验在几个频谱拍卖域，我们表明，它实现了高的分配效率，即使只有很少的价值是从投标人。
Combinatorial auctions (CAs) are used to allocate multiple items among bidders with complex valuations. Since the value space grows exponentially in the number of items, it is impossible for bidders to report their full value function even in medium-sized settings. Prior work has shown that current designs often fail to elicit the most relevant values of the bidders, thus leading to inefficiencies. We address this problem by introducing a machine learning-based elicitation algorithm to identify which values to query from the bidders. Based on this elicitation paradigm we design a new CA mechanism we call PVM, where payments are determined so that bidders’ incentives are aligned with allocative efficiency. We validate PVM experimentally in several spectrum auction domains, and we show that it achieves high allocative efficiency even when only few values are elicited from the bidders.