Abu Kaisar Mohammad Masum

Deterministic Hyperdimensional Learning with Rank Refinement (Student Abstract)

Hyperdimensional Computing (HDC) represents data as high-dimensional hypervectors that are robust and efficient for learning. Existing methods often rely on pseudo-random hypervector generation, which can suffer from poor orthogonality and high variance across runs, ultimately slowing convergence. T