Trimmed Density Ratio Estimation

Authors: Song Liu, Akiko Takeda, Taiji Suzuki, Kenji Fukumizu

NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments are conducted to verify the effectiveness of the estimator. and 6 Experiments
Researcher Affiliation Academia Song Liu University of Bristol song.liu@bristol.ac.uk Akiko Takeda The Institute of Statistical Mathematics, AIP, RIKEN, atakeda@ism.ac.jp Taiji Suzuki University of Tokyo, Sakigake (PRESTO), JST, AIP, RIKEN, taiji@mist.i.u-tokyo.ac.jp Kenji Fukumizu The Institute of Statistical Mathematics, fukumizu@ism.ac.jp
Pseudocode Yes Algorithm 1 Gradient Ascent and Trimming
Open Source Code Yes Code can be found at http://allmodelsarewrong.org/software.html
Open Datasets No No specific link, DOI, repository name, or formal citation to a publicly available or open dataset was provided. The datasets described appear to be custom-collected for the experiments, e.g., 'We collect four images (see Figure 3a)...'
Dataset Splits No No specific details on dataset splits (e.g., percentages, sample counts for training, validation, or test sets, or citations to standard splits) were found.
Hardware Specification No No specific hardware details (like GPU models, CPU models, or memory specifications) used for running experiments were mentioned. A general mention of 'GPU acceleration' does not suffice.
Software Dependencies No The paper mentions 'Tensorflow2' but does not provide a specific version number or any other software dependencies with version information.
Experiment Setup Yes To induce sparsity, we set R( ) = Pd i,j=1,i j | i,j| and fix λ = 0.0938. Then run DRE and TRimmed-DRE to learn the sparse differential precision matrix... We fix ν in TR-DRE to 90% and compare the performance of DRE and TR-DRE...