Completing State Representations using Spectral Learning

Authors: Nan Jiang, Alex Kulesza, Satinder Singh

NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Empirical evaluation on synthetic HMMs, an aircraft identification domain, and a gene splice dataset shows that, even with weak domain knowledge, the algorithm can significantly outperform standard PSR learning.
Researcher Affiliation Collaboration UIUC Urbana, IL nanjiang@illinois.edu Alex Kulesza Google Research New York, NY kulesza@google.com Satinder Singh University of Michigan Ann Arbor, MI baveja@umich.edu
Pseudocode Yes Algorithm 1 Template for learning transformed PSR-fs; Algorithm 2 A basic spectral algorithm for PSR-f; Algorithm 3 Canonical angle algorithm for PSR-f
Open Source Code No The paper does not provide any concrete access to source code for the methodology, such as a repository link, or an explicit statement about code release in supplementary materials.
Open Datasets Yes Finally, we experiment on a gene splice dataset [13]. ... [13] Dua Dheeru and EfiKarra Taniskidou. UCI machine learning repository, 2017. URL http://archive.ics.uci.edu/ml.
Dataset Splits Yes The hyperparameter d for Algorithm 3 is tuned by 3-fold cross validation on training data, and λ is set to 100 to ensure a succinct model (see Appendix E.1).
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details, such as library names with version numbers, needed to replicate the experiment.
Experiment Setup Yes The hyperparameter d for Algorithm 3 is tuned by 3-fold cross validation on training data, and λ is set to 100 to ensure a succinct model (see Appendix E.1).