Abstract Machine learning has the potential to significantly improve computer systems. While recent research in this area has shown great promise, not all problems are equally well-suited for applying ML techniques, and some remaining challenges have prevented wider adoption of ML in systems. In this talk, I will introduce a taxonomy to classify machine learning for systems approaches, discuss how to identify cases that are a good fit for machine learning, and lay out a longer-term vision of how systems can be improved using ML techniques, ranging from computer architecture to language runtimes. I will then cover two specific applications of ML for Systems: learning-based memory allocation for C++ and ML for storage systems.

Martin Maas is a Staff Research Scientist at Google DeepMind. His research interests are in language runtimes, computer architecture, systems, and machine learning, with a focus on applying machine learning to systems problems. He also chairs the J Extension group within the RISC-V project, which investigates managed-runtime extensions. Before joining Google, Martin completed his PhD in Computer Science at the University of California at Berkeley, where he worked on hardware support for managed languages and architectural support for memory-trace obliviousness.