A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
We adapted the Covertype data set for unsupervised learning.
Do two data samples come from different distributions? Recent studies of this fundamental problem focused on embedding probability distributions into sufficiently rich characteristic Reproducing Kernel Hilbert Spaces (RKHSs), to compare distributions by the distance between their embeddings. We show that Regularized Ma…
Paper tackles Byzantine attacks in distributed learning with a new ADMM method.
Abc-boost is a new line of boosting algorithms for multi-class classification, by utilizing the commonly used sum-to-zero constraint. To implement abc-boost, a base class must be identified at each boosting step. Prior studies used a very expensive procedure based on exhaustive search for determining the base class at …
Signed Evidence Flow (SEF) combines fitted prediction with signed feature attributions to measure evidence conflict and stability.
A new method for multi-expert learning-to-defer avoids optimization issues.