scispace - formally typeset
Search or ask a question
Topic

Probabilistic latent semantic analysis

About: Probabilistic latent semantic analysis is a research topic. Over the lifetime, 2884 publications have been published within this topic receiving 198341 citations. The topic is also known as: PLSA.


Papers
More filters
Journal ArticleDOI
TL;DR: This work proposes TINA, a correlation learning method by Adaptive Hierarchical Semantic Aggregation that outperforms state-of-the-art, and achieves better adaptation to the multi-level semantic relation and content divergence.
Abstract: With the explosive growth of web data, effective and efficient technologies are in urgent need for retrieving semantically relevant contents of heterogeneous modalities. Previous studies devote efforts to modeling simple cross-modal statistical dependencies, and globally projecting the heterogeneous modalities into a measurable subspace. However, global projections cannot appropriately adapt to diverse contents, and the naturally existing multilevel semantic relation in web data is ignored. We study the problem of semantic coherent retrieval, where documents from different modalities should be ranked by the semantic relevance to the query. Accordingly, we propose TINA, a correlation learning method by adaptive hierarchical semantic aggregation. First, by joint modeling of content and ontology similarities, we build a semantic hierarchy to measure multilevel semantic relevance. Second, with a set of local linear projections and probabilistic membership functions, we propose two paradigms for local expert aggregation, i.e., local projection aggregation and local distance aggregation. To learn the cross-modal projections, we optimize the structure risk objective function that involves semantic coherence measurement, local projection consistency, and the complexity penalty of local projections. Compared to existing approaches, a better bias-variance tradeoff is achieved by TINA in real-world cross-modal correlation learning tasks. Extensive experiments on widely used NUS-WIDE and ICML-Challenge for image–text retrieval demonstrate that TINA better adapts to the multilevel semantic relation and content divergence, and, thus, outperforms state of the art with better semantic coherence.

52 citations

Journal ArticleDOI
TL;DR: A model that learns semantic representations from the distributional statistics of language, and infers semantic representations by taking into account the inherent sequential nature of linguistic data is described.
Abstract: In this paper, we describe a model that learns semantic representations from the distributional statistics of language. This model, however, goes beyond the common bag-of-words paradigm, and infers semantic representations by taking into account the inherent sequential nature of linguistic data. The model we describe, which we refer to as a Hidden Markov Topics model, is a natural extension of the current state of the art in Bayesian bag-of-words models, that is, the Topics model of Griffiths, Steyvers, and Tenenbaum (2007), preserving its strengths while extending its scope to incorporate more fine-grained linguistic information.

51 citations

Posted Content
TL;DR: The capability of the convolutional VAE model to modify the phonetic content or the speaker identity for speech segments using the derived operations, without the need for parallel supervisory data is demonstrated.
Abstract: An ability to model a generative process and learn a latent representation for speech in an unsupervised fashion will be crucial to process vast quantities of unlabelled speech data. Recently, deep probabilistic generative models such as Variational Autoencoders (VAEs) have achieved tremendous success in modeling natural images. In this paper, we apply a convolutional VAE to model the generative process of natural speech. We derive latent space arithmetic operations to disentangle learned latent representations. We demonstrate the capability of our model to modify the phonetic content or the speaker identity for speech segments using the derived operations, without the need for parallel supervisory data.

51 citations

Journal ArticleDOI
TL;DR: In this paper, the authors describe the general time-intensive longitudinal latent class modeling framework implemented in Mplus and discuss the Bayesian estimation based on Markov chain Monto Carlo, which allows modeling with arbitrary long time series data and many random effects.
Abstract: This article describes the general time-intensive longitudinal latent class modeling framework implemented in Mplus. For each individual a latent class variable is measured at each time point and the latent class changes across time follow a Markov process (i.e., a hidden or latent Markov model), with subject-specific transition probabilities that are estimated as random effects. Such a model for single-subject data has been referred to as the regime-switching state-space model. The latent class variable can be measured by continuous or categorical indicators, under the local independence condition, or more generally by a class-specific structural equation model or a dynamic structural equation model. We discuss the Bayesian estimation based on Markov chain Monto Carlo, which allows modeling with arbitrary long time series data and many random effects. The modeling framework is illustrated with several simulation studies.

51 citations

Proceedings Article
Philip Bachman1
01 Jan 2016
TL;DR: In this article, a lightweight autoregressive model is incorporated in the reconstruction distribution to enable end-to-end training of models with 10+ layers of latent variables, which achieves state-of-the-art performance on standard image modelling benchmarks.
Abstract: We present an architecture which lets us train deep, directed generative models with many layers of latent variables. We include deterministic paths between all latent variables and the generated output, and provide a richer set of connections between computations for inference and generation, which enables more effective communication of information throughout the model during training. To improve performance on natural images, we incorporate a lightweight autoregressive model in the reconstruction distribution. These techniques permit end-to-end training of models with 10+ layers of latent variables. Experiments show that our approach achieves state-of-the-art performance on standard image modelling benchmarks, can expose latent class structure in the absence of label information, and can provide convincing imputations of occluded regions in natural images.

51 citations


Network Information
Related Topics (5)
Feature extraction
111.8K papers, 2.1M citations
84% related
Feature (computer vision)
128.2K papers, 1.7M citations
84% related
Support vector machine
73.6K papers, 1.7M citations
84% related
Deep learning
79.8K papers, 2.1M citations
83% related
Object detection
46.1K papers, 1.3M citations
82% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202319
202277
202114
202036
201927
201858