Conformalized Deep Splines for Optimal and Efficient Prediction Sets. (arXiv:2311.00774v1 [cs.LG])

Uncertainty estimation is critical in high-stakes machine learning
applications. One effective way to estimate uncertainty is conformal
prediction, which can provide predictive inference with statistical coverage
guarantees. We present a new conformal regression method, Spline Prediction
Intervals via Conformal Estimation (SPICE), that estimates the conditional
density using neural-network-parameterized splines. We prove universal
approximation and optimality results for SPICE, which are empirically validated
by our experiments. SPICE is compatible with two different efficient-to-compute
conformal scores, one oracle-optimal for marginal coverage (SPICE-ND) …

applications arxiv inference machine machine learning machine learning applications network prediction predictive regression spline statistical uncertainty