Stochastic Gradient MCMC Methods for Hidden Markov Models
Authors: Yi-An Ma, Nicholas J. Foti, Emily B. Fox
ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the effectiveness of our algorithm on synthetic experiments and an ion channel recording data, with runtimes significantly outperforming batch MCMC. |
| Researcher Affiliation | Academia | 1University of Washington, Seattle, WA, USA. |
| Pseudocode | Yes | Algorithm 1 SG-MCMC [...] Algorithm 2 SG-MCMC for HMM |
| 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 | We evaluate the performance of our proposed SG-RLD algorithm for HMMs on both synthetic and real data. [...] Finally, we apply SG-RLD to a large ion channel recording data set and compare to batch MCMC. In particular, we consider a 1MHz recording from Rosenstein et al. (2013) of a single alamethicin channel. This data was previously investigated in Palla et al. (2014) and Tripuraneni et al. (2015)... |
| Dataset Splits | No | The paper mentions the use of synthetic and real datasets, and refers to "held out probability" and "test data" but does not specify explicit training, validation, or test splits by percentages or sample counts. For example, it does not state "70% for training, 15% for validation, 15% for testing" or similar. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory, cloud instance types) used to run the experiments. |
| Software Dependencies | No | The paper does not provide specific version numbers for any software libraries, frameworks, or programming languages used in the experiments. |
| Experiment Setup | Yes | Algorithm 2 SG-MCMC for HMM: initialize A(0) and φ(0) k for n = 0, 1, 2 , Niter do Periodically estimate the buffer length B and the minimum subchain gap ν according to Sec. 3. Sample subchains S of length L from p(e S). [...] Further details on these datasets and how we set L and |e S| are in the Supplement. [...] Details of the parameter settings used to generate the data are in the Supplement. |