Adaptive Belief Propagation

Authors: Georgios Papachristoudis, John Fisher

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

Reproducibility Variable Result LLM Response
Research Type Experimental We show in synthetic and real experiments that it outperforms standard BP by orders of magnitude and explore the settings that it is advantageous over (S umer et al., 2011).
Researcher Affiliation Academia Georgios Papachristoudis GEOPAPA@MIT.EDU John W. Fisher III FISHER@CSAIL.MIT.EDU CSAIL, MIT, Cambridge, MA 02139, USA
Pseudocode Yes Algorithm 1 ADAPTIVE BP
Open Source Code Yes An implementation of this algorithm is publicly available at https://github.com/geopapa11/adabp
Open Datasets Yes Specifically, we explore the effects of pointwise mutations in DNA sequences to the birth or disappearance of Cp G islands...obtained from the NCBI database. We analyzed temperature measurements collected from 53 wireless sensors at 30 sec intervals from the Intel Berkeley Research lab.
Dataset Splits No The paper mentions training models and using various datasets, but it does not specify explicit train/validation/test split percentages or sample counts for reproducibility in the main text.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts) used for running its experiments.
Software Dependencies No The paper mentions tools and other methods used (e.g., 'Cp G Island Searcher', 'RCTree BP') but does not specify software dependencies with version numbers (e.g., Python, PyTorch versions) for reproducibility.
Experiment Setup No The paper describes general experimental procedures and data usage (e.g., 'randomly generate different w orders', 'train the parameters of HMM') but does not provide specific hyperparameters or system-level training settings like learning rates, batch sizes, or optimizer details in the main text.