2021/06/25 by Anastasios Zouzias, Zouzias, Anastasios, Kleovoulos Kalaitzidis +3 · 1 citation
Computer Science · Psychology · #Artificial Intelligence (cs.AI) #Artificial intelligence #Computer science #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine learning #Psychology #Reinforcement #Reinforcement Learning in Robotics #Reinforcement learning #Social psychology #Software Engineering Research #cs.AI #cs.LG
paper · pdf · doi:10.48550/arxiv.2106.13429
6 pages, appeared in ML workshop for Computer Architecture and Systems 2021
arxiv created 2021/06/25 · openalex publication_date 2021/06/25 · arxiv updated 2021/06/28 · openalex created_date 2022/07/25 · openalex updated_date 2026/08/05
Recent years have seen stagnating improvements to branch predictor (BP) efficacy and a dearth of fresh ideas in branch predictor design, calling for fresh thinking in this area. This paper argues that looking at BP from the viewpoint of Reinforcement Learning (RL) facilitates systematic reasoning about, and exploration of, BP designs. We describe how to apply the RL formulation to branch predictors, show that existing predictors can be succinctly expressed in this formulation, and study two RL-based variants of conventional BPs.