Variational Encoder-based Reliable Classification
This addresses reliability issues in ML predictions for applications requiring trustworthy decisions, though it appears incremental as it builds on existing variational auto-encoder methods.
The paper tackles the problem of unreliable individual predictions in machine learning by proposing an Epistemic Classifier (EC) that uses modified variational auto-encoders to provide justification based on training data support and reconstruction quality, resulting in improved reliability and robust identification of adversarial attacks compared to softmax-based thresholding.
Machine learning models provide statistically impressive results which might be individually unreliable. To provide reliability, we propose an Epistemic Classifier (EC) that can provide justification of its belief using support from the training dataset as well as quality of reconstruction. Our approach is based on modified variational auto-encoders that can identify a semantically meaningful low-dimensional space where perceptually similar instances are close in $\ell_2$-distance too. Our results demonstrate improved reliability of predictions and robust identification of samples with adversarial attacks as compared to baseline of softmax-based thresholding.