Adaptive Classification for Prediction Under a Budget

TL;DR

This paper introduces an adaptive prediction method that dynamically selects models at test time to minimize cost while maintaining high accuracy, using a bottom-up training strategy.

Contribution

It presents a novel bottom-up approach to jointly train gating and prediction models for cost-effective, adaptive prediction under resource constraints.

Findings

Outperforms state-of-the-art methods on benchmark datasets

Achieves higher accuracy at the same cost levels

Effectively adapts model complexity to input regions

Abstract

We propose a novel adaptive approximation approach for test-time resource-constrained prediction. Given an input instance at test-time, a gating function identifies a prediction model for the input among a collection of models. Our objective is to minimize overall average cost without sacrificing accuracy. We learn gating and prediction models on fully labeled training data by means of a bottom-up strategy. Our novel bottom-up method first trains a high-accuracy complex model. Then a low-complexity gating and prediction model are subsequently learned to adaptively approximate the high-accuracy model in regions where low-cost models are capable of making highly accurate predictions. We pose an empirical loss minimization problem with cost constraints to jointly train gating and prediction models. On a number of benchmark datasets our method outperforms state-of-the-art achieving higher accuracy for the same cost.