Abstract: The challenge in the widely applicable online matching problem lies in making irrevocable assignments while there is uncertainty about future inputs. Most theoretically-grounded policies are myopic or greedy in nature. In real-world applications where the matching process is repeated on a regular basis, the underlying data distribution can be leveraged for better decision-making. We present an end-to-end Reinforcement Learning framework for deriving better matching policies based on trial-and-error on historical data. We devise a set of neural network architectures, design feature representations, and empirically evaluate them across two online matching problems: Edge-Weighted Online Bipartite Matching and Online Submodular Bipartite Matching. We show that most of the learning approaches perform consistently better than classical baseline algorithms on four synthetic and real-world datasets. On average, our proposed models improve the matching quality by 3-10% on a variety of synthetic and real-world datasets.
License: Creative Commons Attribution 4.0 International (CC BY 4.0)
Submission Length: Long submission (more than 12 pages of main content)
Changes Since Last Submission: The camera ready version has not gone through any changes from the previous submission - acceptance date, author names and OpenReview link added.
Assigned Action Editor: ~Christian_Kroer1
Submission Number: 220