Deep Weighted Averaging Classifiers
This addresses concerns about the reliability and interpretability of deep learning models for users in fields requiring trustworthy AI, though it is incremental as it builds on existing conformal methods.
The paper tackled the issues of calibration, robustness, and interpretability in deep learning classifiers by proposing a method that modifies deep architectures to provide transparent explanations and credibility measures for predictions, resulting in improved transparency, controlled error rates, and robustness without compromising accuracy or calibration.
Recent advances in deep learning have achieved impressive gains in classification accuracy on a variety of types of data, including images and text. Despite these gains, however, concerns have been raised about the calibration, robustness, and interpretability of these models. In this paper we propose a simple way to modify any conventional deep architecture to automatically provide more transparent explanations for classification decisions, as well as an intuitive notion of the credibility of each prediction. Specifically, we draw on ideas from nonparametric kernel regression, and propose to predict labels based on a weighted sum of training instances, where the weights are determined by distance in a learned instance-embedding space. Working within the framework of conformal methods, we propose a new measure of nonconformity suggested by our model, and experimentally validate the accompanying theoretical expectations, demonstrating improved transparency, controlled error rates, and robustness to out-of-domain data, without compromising on accuracy or calibration.