Oct 21, 2020 · We propose an end-to-end, transferable deep reinforcement learning method for computational graph optimization (GO), based on a scalable sequential attention ...
scholar.google.com › citations
We consider three related computational graph optimization problems from the ML compiler optimization stack. We first explain how each of these problems can be ...
Most compilers for machine learning (ML) frameworks need to solve many correlated optimization problems to generate efficient machine code.
Summary and Contributions: This paper presents GO, a reinforcement learning method for optimizing execution of computation graphs.
This work proposes an end-to-end, transferable deep reinforcement learning method for computational graph optimization (GO), based on a scalable sequential ...
Feb 20, 2021 · In mapping a computational graph to machine code that executes on a collection of devices, ML compilers need to solve many optimization problems ...
Generalises across different graphs and tasks – move varied set. • Work on entire graph at once instead of just one node at a time – capture long distance.
May 25, 2024 · Would you recommend learning topics like IR optimization, register/memory allocation or SSA first before getting into ML compilers?
Papertalk is an open-source platform where scientists share video presentations about their newest scientific results - and watch, like + discuss them.
Most compilers for machine learning (ML) frameworks need to solve many correlated optimization problems to generate efficient machine code.