Generative and Multi-phase Learning for Computer Systems Optimization (ISCA 2019)
and
AutoTap: Synthesizing and Repairing Trigger-Action Programs Using LTL Properties
This week’s Pizza Seminar will be a double feature, with two conference talks presented by Yi Ding (@dingy) and Lefan Zhang (@lefanz). Twice as many speakers, twice as much reason to come! The abstracts are below,
Generative and Multi-phase Learning for Computer Systems Optimization (Yi Ding, ISCA 2019)
Machine learning and artificial intelligence are invaluable for computer systems optimization: as computer systems expose more resources for management, ML/AI is necessary for modeling these resources’ complex interactions. The standard way to incorporate ML/AI into a computer system is to first train a learner to accurately predict the system’s behavior as a function of resource usage and then deploy the learned model as part of a system. In this talk, I’m going to show that (1) continued improvement of learning accuracy may not improve the systems result, but (2) incorporating knowledge of the systems problem into the learning process improves the systems results even though it may not improve overall accuracy. Our results imply that learning for systems optimization may have reached a point of diminishing returns where accuracy improvements have little effect on the systems outcome. Thus we advocate that future work on learning for systems should de-emphasize accuracy and instead incorporate the system problem’s structure into the learner.
AutoTap: Synthesizing and Repairing Trigger-Action Programs Using LTL Properties (Lefan Zhang)
End-user programming, particularly trigger-action programming (TAP), is a popular method of letting users express their intent for how smart devices and cloud services interact. Unfortunately, in some situations it can be challenging for users to correctly express their desires through TAP. This paper presents AutoTap, a system that lets novice users easily specify desired properties for devices and services. AutoTap translates these properties to linear temporal logic (LTL) and both automatically synthesizes property-satisfying TAP rules from scratch and repairs existing TAP rules. We designed AutoTap based on a user study mapping the properties users wish to express. Through a second user study, we show that novice users are significantly more likely to express some desired behaviors correctly using AutoTap than using TAP rules. From our benchmarks and experiments, we find AutoTap is a simple and effective option for correct and expressive end-user programming.
Adding here for the sake of completeness of pizza