Abstract

Machine Learning (ML) research has focused on maximizing the accuracy of predictive tasks. ML models, however, are increasingly more complex, resource intensive, and costlier to deploy in resource-constrained environments. These issues are exacerbated for prediction tasks with sequential classification on progressively transitioned stages with “happens-before” relation between them.

We argue that it is possible to “unfold” a monolithic single multi-class classifier, typically trained for all stages using all data, into a series of single-stage classifiers. Each single-stage classifier can be cascaded gradually from cheaper to more expensive binary classifiers that are trained using only the necessary data modalities or features required for that stage.

Key Result

UnfoldML is a cost-aware and uncertainty-based dynamic 2D prediction pipeline for multi-stage classification that enables:

1. Navigation of the accuracy-cost tradeoff space
2. Reduction of spatio-temporal inference cost by orders of magnitude
3. Earlier prediction on proceeding stages

UnfoldML achieves within 0.1% accuracy of the highest-performing multi-class baseline in clinical settings, while saving close to 20× on spatio-temporal inference cost and predicting disease onset 3.5 hours earlier. The framework also generalizes to image classification, saving close to 5× cost with as little as 0.4% accuracy reduction.

Cite This Paper

@inproceedings{xu2022unfoldml,
  title     = {UnfoldML: Cost-Aware and Uncertainty-Based Dynamic 2D Prediction for Multi-Stage Classification},
  author    = {Xu, Yanbo and Khare, Alind and Matlin, Glenn and Ramadoss, Monish and Kamaleswaran, Rishikesan and Zhang, Chao and Tumanov, Alexey},
  booktitle = {Advances in Neural Information Processing Systems 35 (NeurIPS 2022)},
  pages     = {4598--4611},
  year      = {2022}
}