Macro Mondrian Conformal Risk Control
This paper extends conformal prediction to control general monotone loss functions and addresses imbalanced distributions.
This paper introduces a new method to control risk on average over user-defined groups, which is novel compared to prior work that only guarantees marginal control.
Before reading this…
Applications
- →Semantic segmentation
To understand this paper, make sure you know these concepts first:
- Conformal predictionfind papers →
- Mondrian conformal predictionfind papers →
- Macro-coverage algorithmsfind papers →
Abstract
More Like ThisConformal Risk Control (CRC) extends conformal prediction beyond the coverage setting to the control of general monotone loss functions, in a distribution-free and modelagnostic fashion. This makes it well-suited to problems with high-dimensional targets, such as the semantic segmentation task addressed in this work. However, CRC guarantees only marginal control, meaning that the risk is controlled on average over the data distribution. Under imbalanced distributions, the risk on inputs that contain rare objects may therefore remain uncontrolled, even when the marginal guarantee holds. We propose a new method that controls the risk on average over user-defined groups of the domain, drawing on Mondrian conformal prediction and macro-coverage algorithms. The method relies on a taxonomy, that is, a function that assigns a group to each data point, and reweights each point's risk by a coefficient that depends on its group probability, so that rare points contribute more than common ones. We illustrate the approach on two semantic segmentation datasets.