A Case for Rejection in Low Resource ML Deployment
This work addresses the challenge of deploying AI in resource-limited or early-stage applications, though it appears incremental as it builds on existing sample rejection methods.
The paper tackles the problem of building reliable AI decision support systems in low-resource settings by proposing a simple sample rejection method as a proof of concept baseline, addressing the inadequacy of existing approaches for such scenarios.
Building reliable AI decision support systems requires a robust set of data on which to train models; both with respect to quantity and diversity. Obtaining such datasets can be difficult in resource limited settings, or for applications in early stages of deployment. Sample rejection is one way to work around this challenge, however much of the existing work in this area is ill-suited for such scenarios. This paper substantiates that position and proposes a simple solution as a proof of concept baseline.