e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación
Artículo científico (JournalArticle)
Multinomial sampling of latent variables for hierarchical change-point detection
Bayesian change-point detection, with latent variable models, allows to perform segmentation of high-dimensional time-series with heterogeneous statistical nature. We assume that change-points lie on a lower-dimensional manifold where we aim to infer a discrete representation via subsets of latent variables. For this particular model, full inference is computationally unfeasible and pseudo-observations based on point-estimates of latent variables are used instead. However, if their estimation is not certain enough, change-point detection gets affected. To circumvent this problem, we propose a multinomial sampling methodology that improves the detection rate and reduces the delay while keeping complexity stable and inference analytically tractable. Our experiments show results that outperform the baseline method and we also provide an example oriented to a human behavioral study., This work was supported by the Ministerio de Ciencia, Innovación y Universidades under grant TEC2017-92552-EXP (aMBITION), by the Ministerio de Ciencia, Innovación y Universidades, jointly with the European Commission (ERDF), under grants TEC2017-86921-C2-2-R (CAIMAN) and RTI2018-099655-B-I00 (CLARA), and by the Comunidad de Madrid under grant Y2018/TCS-4705 (PRACTICO-CM). The work of PMM has been supported by FPI grant BES-2016-077626 and ERC funding under the EU’s Horizon 2020 research and innovation programme (grant agreement nº 757360). LRM has been supported by grant IND2018/TIC-9649 from the Comunidad de Madrid.
Proyecto:
EC / H2020 757360