Greedy Algorithms for Cone Constrained Optimization with Convergence Guarantees

Authors: Francesco Locatello, Michael Tschannen, Gunnar Raetsch, Martin Jaggi

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

Reproducibility Variable Result LLM Response
Research Type Experimental We illustrate the performance of the presented algorithms on three different exemplary tasks, showing that our algorithms are competitive with established baselines across a wide range of objective functions, domains, and data sets while not being specifically tailored to any of these tasks (see Section 3.2 for a discussion of the computational complexity of the algorithms). Additional experiments targeting KL divergence NMF, non-negative tensor factorization, and hyperspectral image unmixing can be found in the appendix.6 Illustrative Experiments
Researcher Affiliation Academia Francesco Locatello MPI for Intelligent Systems ETH Zurich locatelf@ethz.ch Michael Tschannen ETH Zurich michaelt@nari.ee.ethz.ch Gunnar Rätsch ETH Zurich raetsch@inf.ethz.ch Martin Jaggi EPFL martin.jaggi@epfl.ch
Pseudocode Yes Algorithm 1 Norm-Corrective Generalized Matching Pursuit, Algorithm 2 Non-Negative Matching Pursuit, Algorithm 3 Away-steps (AMP) and Pairwise (PWMP) Non-Negative Matching Pursuit, Algorithm 4 Fully Corrective Non-Negative Matching Pursuit (FCMP)
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets Yes We consider three different datasets: The Reuters Corpus2, the CBCL face dataset3 and the KNIX dataset4. 2http://www.nltk.org/book/ch02.html 3http://cbcl.mit.edu/software-datasets/Face Data2.html 4http://www.osirix-viewer.com/resources/dicom-image-library/
Dataset Splits No The paper mentions 'training and test accuracy on 100 random splits' for the non-negative garrote experiment but does not provide specific percentages, sample counts, or explicit validation splits for dataset partitioning.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers (e.g., library or solver names with versions) needed to replicate the experiment.
Experiment Setup No The paper describes the general setup of the experiments and the algorithms being compared but does not specify concrete hyperparameter values or detailed training configurations for its models.