ceClub: Deep Reinforcement Learning in Compiler

Speaker:
Ameer Haj Ali (UC Berkeley)
Date:
Wednesday, 8.1.2020, 15:30
Place:
Electrical Eng. Building 815

Compilers are designed today to use fixedAbstract -cost models that are based on heuristics to make different code optimizations. However, these models are unable to capture the data dependency, the computation graph, or the organization of instructions. Therefore, with these models compilers achieve suboptimal performance. In his talk, Ameer will show how to use deep reinforcement learning to overcome such hard compiler challenges. This includes the NP-Hard compiler phase ordering challenge (AutoPhase, FCCM2019, SysML2020) and automatically tuning compiler pragmas such as vectorization (NeuroVectorizer, CGO2020).

Back to the index of events