Binary adaptive embeddings from order statistics of random projections
This work addresses the need for efficient signal processing in domains like machine learning, but it appears incremental as it builds on existing random projection methods with specific adaptations.
The paper tackles the problem of constructing binary embeddings for signals correlated with a reference signal, using order statistics of random projections, and shows improved performance in classification tasks within a reduced-dimensionality space.
We use some of the largest order statistics of the random projections of a reference signal to construct a binary embedding that is adapted to signals correlated with such signal. The embedding is characterized from the analytical standpoint and shown to provide improved performance on tasks such as classification in a reduced-dimensionality space.