Entropic Graph-based Posterior Regularization

Authors: Maxwell Libbrecht, Michael Hoffman, Jeff Bilmes, William Noble

ICML 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental On a synthetic problem, we show that our method outperforms existing methods for graph-based regularization and a comparable strategy for incorporating long-range interactions using existing methods for approximate inference. Using genome-scale functional genomics data, we integrate genome 3D interaction data into existing models for genome annotation and demonstrate significant improvements in predicting genomic activity. 5. Simulations 6. Application: Genome Annotation Using Physical Interaction Information
Researcher Affiliation Academia Maxwell W Libbrecht MAXWL@CS.WASHINGTON.EDU Genome Sciences, Box 355065, Foege Building, S220B, 3720 15th Ave NE, Seattle, WA 98195-5065 Michael M Hoffman MICHAEL.HOFFMAN@UTORONTO.CA Princess Margaret Cancer Centre, Toronto Medical Discovery Tower 11-311, 101 College St, Toronto, ON M5G 1L7 Jeffrey A Bilmes BILMES@EE.WASHINGTON.EDU Department of Electrical Engineering, University of Washington, Seattle, Box 352500, Seattle, WA 98195-2500 William S Noble WILLIAM-NOBLE@U.WASHINGTON.EDU Genome Sciences, Box 355065, Foege Building, S220B, 3720 15th Ave NE, Seattle, WA 98195-5065
Pseudocode No The full algorithm in pseudocode is shown in the extended version (Libbrecht et al., 2015).
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes On ENCODE data, a model using EGPR predicts genome activity much more accurately than the currently-used chain models as well as other forms of regularizer. and (ENCODE Project Consortium, 2012; Hoffman et al., 2012b), http://www.nature.com/encode and We used the Hi-C data set of (Dixon et al., 2012) and processed the Hi-C data into a matrix of pairwise p-values using the Fit-Hi-C method (Ay et al., 2014a). and To evaluate these annotations, we used the time during the cell cycle at which the DNA is replicated as a gold standard (Woodfine et al., 2004).
Dataset Splits Yes We chose the best-performing hyperparameters for each GPR model using a validation set.
Hardware Specification No The paper does not provide specific hardware details (like GPU/CPU models or memory amounts) used for running its experiments.
Software Dependencies No All implementations in the below use the GMTK, the graphical models toolkit (?), and its use of virtual evidence factors, for HMM and dynamic graphical model inference. The paper does not provide specific version numbers for GMTK or other software dependencies.
Experiment Setup No The paper mentions that hyperparameters were chosen using training and validation sets, but it does not specify the concrete values of these hyperparameters (λG, λR1, λR2, λ) or other experimental setup details like learning rates or batch sizes.