Importance Sampling with Unequal Support
This addresses a bottleneck in off-policy evaluation for machine learning, particularly in healthcare applications like diabetes treatment, but is incremental as it builds on existing importance sampling methods.
The paper tackles the problem of high variance in importance sampling when training and testing distributions have different supports, proposing a new estimator that reduces variance by orders of magnitude. It demonstrates this improvement in a diabetes treatment policy example.
Importance sampling is often used in machine learning when training and testing data come from different distributions. In this paper we propose a new variant of importance sampling that can reduce the variance of importance sampling-based estimates by orders of magnitude when the supports of the training and testing distributions differ. After motivating and presenting our new importance sampling estimator, we provide a detailed theoretical analysis that characterizes both its bias and variance relative to the ordinary importance sampling estimator (in various settings, which include cases where ordinary importance sampling is biased, while our new estimator is not, and vice versa). We conclude with an example of how our new importance sampling estimator can be used to improve estimates of how well a new treatment policy for diabetes will work for an individual, using only data from when the individual used a previous treatment policy.