A reversible infinite HMM using normalised random measures

Authors: David Knowles, Zoubin Ghahramani, Konstantina Palla

ICML 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We use our process to construct a reversible infinite HMM which we apply to two real datasets, one from epigenomics and one ion channel recording.In this section we evaluate the SHGP by running the SHGP Hidden Markov model on two real world datasets.
Researcher Affiliation Academia Konstantina Palla KP376@CAM.AC.UK University of Cambridge, Trumpington Street, CB2 1PZ David A. Knowles DAVIDKNOWLES@CS.STANFORD.EDU Stanford University, 353 Serra Mall, CA 94305-9025 Zoubin Ghahramani ZOUBIN@ENG.CAM.AC.UK University of Cambridge, Trumpington Street, CB2 1PZ
Pseudocode No The paper describes the inference steps in paragraph form and bullet points but does not contain formally structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code (specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described in this paper.
Open Datasets Yes For this experiment we used Ch IP-seq (chromatin immunoprecipitation sequencing) data... (see Park (2009) for a nice review).and analysing a 1MHz recording from the state-of-the-art method of Rosenstein et al. (2013) of a single alamethicin channel.
Dataset Splits Yes We ran 10 repeats, each time holding out different 20% of the data Y and using different random initilisation.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper mentions several algorithms and methods used (e.g., slice sampling, HMC, NUTS, beam sampler, forward-backward algorithm) but does not provide specific software names with version numbers or dependencies for replication.
Experiment Setup Yes We used a truncation level of K = 20, 1000 iterations and a burnin of 700. and For SHGP-HMM we used a truncation of K = 15, 1000 iterations and a burnin of 700. 50 inner iterations of HMC or NUTS were run per outer iteration.