TY - RPRT TI - Further and stronger analogy between sampling and optimization: Langevin Monte Carlo and gradient descent AU - Arnak S. Dalalyan PY - 2017 UR - https://arxiv.org/abs/1704.04752 ID - 1704.04752 ER -