Difference between revisions of "Research questions"
From MAGEEC
(Created page with "This page covers some research questions that would be interesting to explore, once we have an initial framework to play with ==Dynamic Features== *Can we better predict whi...") |
(→Feature Vector) |
||
| (3 intermediate revisions by one other user not shown) | |||
| Line 13: | Line 13: | ||
==Interactivity== | ==Interactivity== | ||
*What is the effect of applying N learnt optimizations, and then retaking the features? | *What is the effect of applying N learnt optimizations, and then retaking the features? | ||
| + | |||
| + | ==Feature Vector== | ||
| + | *Which features should be in the feature vector? | ||
| + | Principle Component Analysis has been used to evaluate how much each feature varies across BEEBS. See blog post here: http://mageec.org/2014/08/11/evaluation-of-static-program-features-used-in-mageec/#more-408 | ||
| + | *Are the features compiler specific? | ||
| + | |||
| + | ==Machine Learning== | ||
| + | *What types of machine learning performs best for learning optimizations? | ||
| + | *Can the machine learning learn when to 'backtrack' and undo a previously applied optimization, based on benchmark features? | ||
| + | *How varied a set of benchmarks is needed to properly train a database? | ||
| + | *Can fewer benchmarks be used, but each benchmark altered by applying a random set of transformations? | ||
| + | |||
| + | ==Data Dependence== | ||
| + | *Do different sequences of optimizations need to be applied for different data sets? | ||
| + | |||
| + | ==Multidimensional Cost Functions== | ||
| + | *Can we optimize for energy and performance simultaneously? | ||
| + | *Can the balance between different cost metrics be altered? | ||
| + | *Does the database need to be retrained for different target metrics? | ||
Latest revision as of 13:15, 14 August 2014
This page covers some research questions that would be interesting to explore, once we have an initial framework to play with
Contents
Dynamic Features
- Can we better predict which optimizations to use if we take dynamic features of the application?
- What kinds of dynamic features can we capture?
- GCC/LLVM both have profile guided optimization, can we use this file for the dynamic features?
- Can hardware counters be used?
Non-binary Parameters
- Some optimizations have parameters which aren't on or off. Can be learn good values for these parameters?
Interactivity
- What is the effect of applying N learnt optimizations, and then retaking the features?
Feature Vector
- Which features should be in the feature vector?
Principle Component Analysis has been used to evaluate how much each feature varies across BEEBS. See blog post here: http://mageec.org/2014/08/11/evaluation-of-static-program-features-used-in-mageec/#more-408
- Are the features compiler specific?
Machine Learning
- What types of machine learning performs best for learning optimizations?
- Can the machine learning learn when to 'backtrack' and undo a previously applied optimization, based on benchmark features?
- How varied a set of benchmarks is needed to properly train a database?
- Can fewer benchmarks be used, but each benchmark altered by applying a random set of transformations?
Data Dependence
- Do different sequences of optimizations need to be applied for different data sets?
Multidimensional Cost Functions
- Can we optimize for energy and performance simultaneously?
- Can the balance between different cost metrics be altered?
- Does the database need to be retrained for different target metrics?
