창시자분들)가 분산 머신러닝 훈련 & 강화학습을 위한 연구하던 중 ML을 위한 분산 시스템 개발이 쉽지 않다는걸 깨달음 ML 분산 시스템 개발을 더 쉽게 만들기 위해 Ray가 개발됨 1. TRPO: Monotone Policy Improvement for Deep RL (ICML 2015) 2. GAE: Variance Reduction Techniques for Speeding Up Learning 3. SparkNet: Communication Efficient Deep Learning on Spark Trust Region Policy Optimization John Schulman, Sergey Levine, Philipp Moritz, Michael I. Jordan, Pieter Abbeel (ICML, 2015) Generalized Advantage Estimation John Schulman, Philipp Moritz, Sergey Levine, Michael I. Jordan, Pieter Abbeel (ICLR, 2016) SparkNet: Training Deep Networks in Spark Philipp Moritz, Robert Nishihara, Ion Stoica, Michael I. Jordan (ICLR, 2016)