Perceptual adjustment queries and an inverted measurement paradigm for low-rank metric learning
This addresses metric learning challenges by providing a more efficient way to collect human input, though it appears incremental as it builds on existing query paradigms.
The paper tackles the problem of learning an unknown Mahalanobis distance by introducing perceptual adjustment queries (PAQs), a new human feedback mechanism that is informative and cognitively lightweight, and develops a two-stage estimator with sample complexity guarantees, showing performance in numerical simulations.
We introduce a new type of query mechanism for collecting human feedback, called the perceptual adjustment query ( PAQ). Being both informative and cognitively lightweight, the PAQ adopts an inverted measurement scheme, and combines advantages from both cardinal and ordinal queries. We showcase the PAQ in the metric learning problem, where we collect PAQ measurements to learn an unknown Mahalanobis distance. This gives rise to a high-dimensional, low-rank matrix estimation problem to which standard matrix estimators cannot be applied. Consequently, we develop a two-stage estimator for metric learning from PAQs, and provide sample complexity guarantees for this estimator. We present numerical simulations demonstrating the performance of the estimator and its notable properties.