Technique that implicitly maps data into higher dimensions for linear separation via kernel functions. ← Keccak‑256 Nakamoto Consensus →