SDP Relaxation with Randomized Rounding for Energy Disaggregation

Authors: Kiarash Shaloudegi, András György, Csaba Szepesvari, Wilsun Xu

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Simulation results both in synthetic and real-world datasets demonstrate the superiority of our method.
Researcher Affiliation Academia Kiarash Shaloudegi Imperial College London k.shaloudegi16@imperial.ac.uk András György Imperial College London a.gyorgy@imperial.ac.uk Csaba Szepesvári University of Alberta szepesva@ualberta.ca Wilsun Xu University of Alberta wxu@ualberta.ca
Pseudocode Yes Algorithm 1 ADMM-RR: Randomized rounding algorithm for suboptimal solution to (2)
Open Source Code Yes Our code is available online at https://github.com/kiarashshaloudegi/FHMM_inference.
Open Datasets Yes we used the REDD dataset of Kolter and Johnson [2011]
Dataset Splits No We use the first half of the data for training and the second half for testing.
Hardware Specification No The paper mentions memory usage (e.g., '14GB vs 6GB in our case') and notes implementation languages (Matlab, C++) and solvers (MOSEK), but does not specify hardware components like CPU/GPU models or total system memory.
Software Dependencies No The paper mentions using 'MOSEK inside the Matlab-based YALMIP', but does not provide specific version numbers for these software components.
Experiment Setup Yes Our method, shown under the label ADMM-RR, runs ADMM for 2500 iterations, runs the local search at the end of each 250 iterations, and chooses the result that has the maximum likelihood.