Hidden 1-Counter Markov Models and How to Learn Them
Authors: Mehmet Kurucan, Mete Özbaltan, Sven Schewe, Dominik Wojtczak
IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | All algorithms here were implemented in Python and evaluated on Intel i7 3.3 GHz CPU with 16 GB RAM. We also implemented the standard algorithms (forward, backward, Baum-Welch) for HMMs ourselves for a fair comparison. ... We tested the algorithm on the following input. We first created 6 different models... Then using each Hi, we created a random multi-set of observation sequences Oi(Hi, T, 20000)... |
| Researcher Affiliation | Academia | Ardahan University, Ardahan, Turkey 2Erzurum Technical University, Erzurum, Turkey 3University of Liverpool, Liverpool, UK |
| Pseudocode | No | While the paper provides detailed step-by-step mathematical descriptions of the adapted algorithms (Forward, Backward, Baum-Welch) with base and recursion steps, these are presented as formal definitions and equations rather than explicitly labeled 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | No | The full source-code and the inputs used will be made freely available online. This statement indicates future availability rather than concrete access at the time of publication. |
| Open Datasets | No | The paper describes generating its own observation sequences: 'Then using each Hi, we created a random multi-set of observation sequences Oi(Hi, T, 20000) that contain 20000 observation sequences with a fixed length T = 16.' No information about public availability of this generated data, standard public datasets, or access links is provided. |
| Dataset Splits | No | The paper mentions 'a set of test observations, which were different from the observation sequences used in the learning process' implying a train/test split, but no specific percentages, counts, or explicit mention of a validation split is provided. |
| Hardware Specification | Yes | All algorithms here were implemented in Python and evaluated on Intel i7 3.3 GHz CPU with 16 GB RAM. |
| Software Dependencies | No | The paper states 'All algorithms here were implemented in Python' but does not specify the version of Python or any other software dependencies with version numbers. |
| Experiment Setup | Yes | We started with 100 different initial completely random (i.e., fully connected with probabilities picked uniformly at random) H1MM models (or HMM models). After each learning step, we discarded the bottom 25% of these models as measured by the value of their likelihood. Eventually we only had one model left that we trained until the learning process converged. |