Similarity-Distance-Magnitude Activations
This work addresses robustness and interpretability issues in selective classification for machine learning practitioners, representing an incremental improvement over existing methods.
The paper tackles the problem of improving robustness and interpretability in selective classification by introducing the Similarity-Distance-Magnitude (SDM) activation function and estimator, which outperforms existing calibration methods using softmax activations in handling co-variate shifts and out-of-distribution inputs.
We introduce the Similarity-Distance-Magnitude (SDM) activation function, a more robust and interpretable formulation of the standard softmax activation function, adding Similarity (i.e., correctly predicted depth-matches into training) awareness and Distance-to-training-distribution awareness to the existing output Magnitude (i.e., decision-boundary) awareness, and enabling interpretability-by-exemplar via dense matching. We further introduce the SDM estimator, based on a data-driven partitioning of the class-wise empirical CDFs via the SDM activation, to control the class- and prediction-conditional accuracy among selective classifications. When used as the final-layer activation over pre-trained language models for selective classification, the SDM estimator is more robust to co-variate shifts and out-of-distribution inputs than existing calibration methods using softmax activations, while remaining informative over in-distribution data.