To develop an imputation algorithm capable of handling different types of missingness in tabular datasets.

First, the organizers will release the development data consisting of a complete dataset along with three missingness masks meant to simulate different patterns of missing data. Participants are encouraged to split this dataset as they see fit for training and validation.

Finally, two held out test sets of equal size will be released, each with a different, unspecified missingness mechanism. The public leaderboard will include 20% of each test set and the private leaderboard will include the remaining 80% of each test set.

We expect the winning submissions will likely be an algorithm that can automatically characterize the pattern of missing data and adapt its imputation approach to the missingness type.