Research Results

Deterministic Relations - Form-free Dependencies - Performance Models

We demonstrate our research results with examples of performance analysis.

Handling Deterministic Relations/Information equivalences

Technical report The Representation and Learning of Equivalent Information in Causal Models.

For experimental results, see menu Data -> Load experiment -> Deterministic relations
Hint: Choose yes when asked to load the accompanying independence map, otherwise all conditional independencies should be recalculated which can take a (very) long time.

Handling Mixed Data & Form-free Dependency Metric

Draft paper Learning Causal Performance Models.

Accurate Performance Model Construction

More results on the benefits of causal learning in analyzing performance can be found under Jan's publications.

Draft paper Learning Causal Performance Models.

More information: Jan Lemeire (jan.lemeire@vub.ac.be)

last updated: August 2007 by Jan Lemeire
Parallel Computing Lab, VUB