Tutorial on Structured Continuous-Time Markov Processes
Authors: C. R. Shelton, G. Ciardo
JAIR 2014 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | This tutorial is intended for readers interested in learning about continuous-time Markov processes, and in particular compact or structured representations of them. It is assumed that the reader is familiar with general probability and statistics and has some knowledge of discrete-time Markov chains and perhaps hidden Markov model algorithms. While this tutorial deals only with Markovian systems, we do not require that all variables be observed. |
| Researcher Affiliation | Academia | Christian R. Shelton EMAIL University of California, Riverside Gianfranco Ciardo EMAIL Iowa State University |
| Pseudocode | Yes | Figure 10: Pseudo-code for sum of quasi-reduced EV MDDs (a and b are either Ωor nodes at level k). The fully-reduced version is similar but slightly more involved, as it needs to take into account the levels of a and b. Figure 12: A Jacobi-style iteration for the stationary solution (πQ = dπdt = 0) when R (non-diagonal elements of Q) is stored as a monolithic EV MDD and h (inverse absolute value of the diagonal elements of Q) is stored in a vector. σ,r is the encoding of R and 0,p is the encoding of the mapping from states to indices (for π and h). Figure 18: Algorithm to sample from CTBN |
| Open Source Code | No | The paper does not provide concrete access to source code. It discusses various existing tools (PRISM, Möbius, SMART) but does not state that the authors are releasing code for the methodology presented in this tutorial. |
| Open Datasets | No | The paper is a tutorial and describes theoretical models and algorithms; it does not present new experimental results with a dataset that the authors are making publicly available. It references applications and studies by other researchers that used specific datasets (e.g., FGF pathway, life event history data, network intrusion detection data), but it does not provide access information for these datasets in the context of the current paper. |
| Dataset Splits | No | The paper is a theoretical tutorial and does not present its own experimental results requiring dataset splits. It discusses methods for parameter estimation and inference but does not specify any training/test/validation splits for a particular dataset. |
| Hardware Specification | No | The paper is a tutorial and primarily focuses on theoretical concepts and algorithms. It does not describe experiments conducted by the authors that would require specific hardware specifications. It mentions 'a modern workstation' generally in the context of problem scalability but not as a specific experimental setup. |
| Software Dependencies | No | The paper is a tutorial and discusses various theoretical models and algorithms. While it mentions several software tools and concepts (e.g., PRISM, Möbius, SMART, Python, PyTorch, CUDA, CPLEX), these are either discussed in a general context, refer to third-party tools, or lack specific version numbers for the authors' own methodology, which is not primarily an implementation paper. |
| Experiment Setup | No | The paper is a theoretical tutorial and does not present specific experimental setups, hyperparameter values, or training configurations from its own work. It describes algorithms and methods but does not detail how one would set up an experiment with specific values for parameters or training processes. |