Research map / Time series / Causal and generative time series
The Segmented iHMM: A Simple, Efficient Hierarchical Infinite HMM
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add state-independent segmentation (change-point) mechanism to iHMM (feature-independent siHMM) | iHMM · Users | labeling error %: 30% -> 25% | Helped |
| add observation features to the siHMM segmentation variable (feature-based) (ablation) | siHMM (feature-independent) · Users | labeling error %: 25% -> 16% | Helped |
| add observation features to the siHMM segmentation variable (feature-based) (ablation) | siHMM (feature-independent) · Drosophila | labeling error %: 37% -> 34% | Helped |
| add observation features to the siHMM segmentation variable (feature-based) (ablation) | siHMM (feature-independent) · Sensors | labeling error %: 18% -> 34% | Hurt |
| add observation features to the siHMM segmentation variable (feature-based) (ablation) | siHMM (feature-independent) · Synthetic | normalized Hamming distance: 0.15 -> 0.13 | Helped |
| use siHMM (feature-independent) instead of iHMM | iHMM · Synthetic | normalized Hamming distance: 0.21 -> 0.15 | Helped |
| use siHMM (feature-independent) instead of iHMM | iHMM · Sensors | labeling error %: 27% -> 18% | Helped |
| use feature-based siHMM instead of iHMM | iHMM · Drosophila | labeling error %: 36% -> 34% | Helped |
| use siHMM (feature-independent) instead of two-level sub-iHMM | sub-iHMM · Users | labeling error %: 24% vs 25% | No clear effect |
| add a state-independent change-point mechanism to the HHMM-style model; compare sub-iHMM baseline | sub-iHMM · Sensors | labeling error %: 22% vs 18% | No clear effect |
Closest papers by meaning
- FlowHMM: Flow-based continuous hidden Markov models (NeurIPS 2022 · similarity 0.88)
- Scaling Factorial Hidden Markov Models: Stochastic Variational Inference without Messages (NeurIPS 2016 · similarity 0.87)
- RED-HDP-HMM: Observation-Dependent Durations for Bayesian Nonparametric Sequential Models (ICML 2026 · similarity 0.86)
- Fast and Consistent Learning of Hidden Markov Models by Incorporating Non-Consecutive Correlations (ICML 2020 · similarity 0.86)
- An Infinite Hidden Markov Model With Similarity-Biased Transitions (ICML 2017 · similarity 0.86)
- LogSig-SSM: Time-Series Modelling with Multi-Scale Log-Signature Compression for State-Space Models (NeurIPS 2026 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.