arXiv subjects

Pablo M. Olmos

Publications and source records attributed to Pablo M. Olmos.

At least 37 records · Page 2Linked to original sources

arXiv2022

Heterogeneous Hidden Markov Models for Sleep Activity Recognition from Multi-Source Passively Sensed Data

Psychiatric patients' passive activity monitoring is crucial to detect behavioural shifts in real-time, comprising a tool that helps clinicians supervise patients' evolution over time and enhance the associated treatments' outcomes. Frequently, sleep disturbances and mental health deterioration are closely related, as mental health condition worsening regularly entails shifts in the patients' circadian rhythms. Therefore, Sleep Activity Recognition constitutes a behavioural marker to portray patients' activity cycles and to detect behavioural changes among them. Moreover, mobile passively sensed data captured from smartphones, thanks to these devices' ubiquity, constitute an excellent alternative to profile patients' biorhythm. In this work, we aim to identify major sleep episodes based on passively sensed data. To do so, a Heterogeneous Hidden Markov Model is proposed to model a discrete latent variable process associated with the Sleep Activity Recognition task in a self-supervised way. We validate our results against sleep metrics reported by clinically tested wearables, proving the effectiveness of the proposed approach.

eess.SP↗

arXiv2022

Multi-task longitudinal forecasting with missing values on Alzheimer's Disease

Machine learning techniques typically applied to dementia forecasting lack in their capabilities to jointly learn several tasks, handle time dependent heterogeneous data and missing values. In this paper, we propose a framework using the recently presented SSHIBA model for jointly learning different tasks on longitudinal data with missing values. The method uses Bayesian variational inference to impute missing values and combine information of several views. This way, we can combine different data-views from different time-points in a common latent space and learn the relations between each time-point while simultaneously modelling and predicting several output variables. We apply this model to predict together diagnosis, ventricle volume, and clinical scores in dementia. The results demonstrate that SSHIBA is capable of learning a good imputation of the missing values and outperforming the baselines while simultaneously predicting three different tasks.

stat.ML↗

arXiv2022

PyHHMM: A Python Library for Heterogeneous Hidden Markov Models

We introduce PyHHMM, an object-oriented open-source Python implementation of Heterogeneous-Hidden Markov Models (HHMMs). In addition to HMM's basic core functionalities, such as different initialization algorithms and classical observations models, i.e., continuous and multinoulli, PyHHMM distinctively emphasizes features not supported in similar available frameworks: a heterogeneous observation model, missing data inference, different model order selection criterias, and semi-supervised training. These characteristics result in a feature-rich implementation for researchers working with sequential data. PyHHMM relies on the numpy, scipy, scikit-learn, and seaborn Python packages, and is distributed under the Apache-2.0 License. PyHHMM's source code is publicly available on Github (https://github.com/fmorenopino/HeterogeneousHMM) to facilitate adoptions and future contributions. A detailed documentation (https://pyhhmm.readthedocs.io/en/latest), which covers examples of use and models' theoretical explanation, is available. The package can be installed through the Python Package Index (PyPI), via 'pip install pyhhmm'.

cs.MS↗

arXiv2021

Deep Autoregressive Models with Spectral Attention

Time series forecasting is an important problem across many domains, playing a crucial role in multiple real-world applications. In this paper, we propose a forecasting architecture that combines deep autoregressive models with a Spectral Attention (SA) module, which merges global and local frequency domain information in the model's embedded space. By characterizing in the spectral domain the embedding of the time series as occurrences of a random process, our method can identify global trends and seasonality patterns. Two spectral attention models, global and local to the time series, integrate this information within the forecast and perform spectral filtering to remove time series's noise. The proposed architecture has a number of useful properties: it can be effectively incorporated into well-know forecast architectures, requiring a low number of parameters and producing interpretable results that improve forecasting accuracy. We test the Spectral Attention Autoregressive Model (SAAM) on several well-know forecast datasets, consistently demonstrating that our model compares favorably to state-of-the-art approaches.

stat.ML↗

arXiv2021

Medical data wrangling with sequential variational autoencoders

Medical data sets are usually corrupted by noise and missing data. These missing patterns are commonly assumed to be completely random, but in medical scenarios, the reality is that these patterns occur in bursts due to sensors that are off for some time or data collected in a misaligned uneven fashion, among other causes. This paper proposes to model medical data records with heterogeneous data types and bursty missing data using sequential variational autoencoders (VAEs). In particular, we propose a new methodology, the Shi-VAE, which extends the capabilities of VAEs to sequential streams of data with missing observations. We compare our model against state-of-the-art solutions in an intensive care unit database (ICU) and a dataset of passive human monitoring. Furthermore, we find that standard error metrics such as RMSE are not conclusive enough to assess temporal models and include in our analysis the cross-correlation between the ground truth and the imputed signal. We show that Shi-VAE achieves the best performance in terms of using both metrics, with lower computational complexity than the GP-VAE model, which is the state-of-the-art method for medical records.

cs.LG↗

arXiv2021

Spatially Coupled Generalized LDPC Codes: Asymptotic Analysis and Finite Length Scaling

Generalized low-density parity-check (GLDPC) codes are a class of LDPC codes in which the standard single parity check (SPC) constraints are replaced by constraints defined by a linear block code. These stronger constraints typically result in improved error floor performance, due to better minimum distance and trapping set properties, at a cost of some increased decoding complexity. In this paper, we study spatially coupled generalized low-density parity-check (SC-GLDPC) codes and present a comprehensive analysis of these codes, including: (1) an iterative decoding threshold analysis of SC-GLDPC code ensembles demonstrating capacity approaching thresholds via the threshold saturation effect; (2) an asymptotic analysis of the minimum distance and free distance properties of SC-GLDPC code ensembles, demonstrating that the ensembles are asymptotically good; and (3) an analysis of the finite-length scaling behavior of both GLDPC block codes and SC-GLDPC codes based on a peeling decoder (PD) operating on a binary erasure channel (BEC). Results are compared to GLDPC block codes, and the advantages and disadvantages of SC-GLDPC codes are discussed.

cs.IT↗

arXiv2021

Bayesian Sparse Factor Analysis with Kernelized Observations

Multi-view problems can be faced with latent variable models since they are able to find low-dimensional projections that fairly capture the correlations among the multiple views that characterise each datum. On the other hand, high-dimensionality and non-linear issues are traditionally handled by kernel methods, inducing a (non)-linear function between the latent projection and the data itself. However, they usually come with scalability issues and exposition to overfitting. Here, we propose merging both approaches into single model so that we can exploit the best features of multi-view latent models and kernel methods and, moreover, overcome their limitations. In particular, we combine probabilistic factor analysis with what we refer to as kernelized observations, in which the model focuses on reconstructing not the data itself, but its relationship with other data points measured by a kernel function. This model can combine several types of views (kernelized or not), and it can handle heterogeneous data and work in semi-supervised settings. Additionally, by including adequate priors, it can provide compact solutions for the kernelized observations -- based in a automatic selection of Bayesian Relevance Vectors (RVs) -- and can include feature selection capabilities. Using several public databases, we demonstrate the potential of our approach (and its extensions) w.r.t. common multi-view learning models such as kernel canonical correlation analysis or manifold relevance determination.

stat.ML↗

arXiv2020

Unsupervised Learning of Global Factors in Deep Generative Models

We present a novel deep generative model based on non i.i.d. variational autoencoders that captures global dependencies among observations in a fully unsupervised fashion. In contrast to the recent semi-supervised alternatives for global modeling in deep generative models, our approach combines a mixture model in the local or data-dependent space and a global Gaussian latent variable, which lead us to obtain three particular insights. First, the induced latent global space captures interpretable disentangled representations with no user-defined regularization in the evidence lower bound (as in $β$-VAE and its generalizations). Second, we show that the model performs domain alignment to find correlations and interpolate between different databases. Finally, we study the ability of the global space to discriminate between groups of observations with non-trivial underlying structures, such as face images with shared attributes or defined sequences of digits images.

cs.LG↗

arXiv2020

Boosting offline handwritten text recognition in historical documents with few labeled lines

In this paper, we face the problem of offline handwritten text recognition (HTR) in historical documents when few labeled samples are available and some of them contain errors in the train set. Three main contributions are developed. First we analyze how to perform transfer learning (TL) from a massive database to a smaller historical database, analyzing which layers of the model need a fine-tuning process. Second, we analyze methods to efficiently combine TL and data augmentation (DA). Finally, an algorithm to mitigate the effects of incorrect labelings in the training set is proposed. The methods are analyzed over the ICFHR 2018 competition database, Washington and Parzival. Combining all these techniques, we demonstrate a remarkable reduction of CER (up to 6% in some cases) in the test set with little complexity overhead.

cs.CV↗

arXiv2020

Improved BiGAN training with marginal likelihood equalization

We propose a novel training procedure for improving the performance of generative adversarial networks (GANs), especially to bidirectional GANs. First, we enforce that the empirical distribution of the inverse inference network matches the prior distribution, which favors the generator network reproducibility on the seen samples. Second, we have found that the marginal log-likelihood of the samples shows a severe overrepresentation of a certain type of samples. To address this issue, we propose to train the bidirectional GAN using a non-uniform sampling for the mini-batch selection, resulting in improved quality and variety in generated samples measured quantitatively and by visual inspection. We illustrate our new procedure with the well-known CIFAR10, Fashion MNIST and CelebA datasets.

stat.ML↗

arXiv2020

Handling Incomplete Heterogeneous Data using VAEs

Variational autoencoders (VAEs), as well as other generative models, have been shown to be efficient and accurate for capturing the latent structure of vast amounts of complex high-dimensional data. However, existing VAEs can still not directly handle data that are heterogenous (mixed continuous and discrete) or incomplete (with missing data at random), which is indeed common in real-world applications. In this paper, we propose a general framework to design VAEs suitable for fitting incomplete heterogenous data. The proposed HI-VAE includes likelihood models for real-valued, positive real valued, interval, categorical, ordinal and count data, and allows accurate estimation (and potentially imputation) of missing data. Furthermore, HI-VAE presents competitive predictive performance in supervised tasks, outperforming supervised models when trained on incomplete data.

cs.LG↗

arXiv2020

Sparse Semi-supervised Heterogeneous Interbattery Bayesian Analysis

The Bayesian approach to feature extraction, known as factor analysis (FA), has been widely studied in machine learning to obtain a latent representation of the data. An adequate selection of the probabilities and priors of these bayesian models allows the model to better adapt to the data nature (i.e. heterogeneity, sparsity), obtaining a more representative latent space. The objective of this article is to propose a general FA framework capable of modelling any problem. To do so, we start from the Bayesian Inter-Battery Factor Analysis (BIBFA) model, enhancing it with new functionalities to be able to work with heterogeneous data, include feature selection, and handle missing values as well as semi-supervised problems. The performance of the proposed model, Sparse Semi-supervised Heterogeneous Interbattery Bayesian Analysis (SSHIBA) has been tested on 4 different scenarios to evaluate each one of its novelties, showing not only a great versatility and an interpretability gain, but also outperforming most of the state-of-the-art algorithms.

stat.ML↗

arXiv2019

Deep Sequential Models for Suicidal Ideation from Multiple Source Data

This article presents a novel method for predicting suicidal ideation from Electronic Health Records (EHR) and Ecological Momentary Assessment (EMA) data using deep sequential models. Both EHR longitudinal data and EMA question forms are defined by asynchronous, variable length, randomly-sampled data sequences. In our method, we model each of them with a Recurrent Neural Network (RNN), and both sequences are aligned by concatenating the hidden state of each of them using temporal marks. Furthermore, we incorporate attention schemes to improve performance in long sequences and time-independent pre-trained schemes to cope with very short sequences. Using a database of 1023 patients, our experimental results show that the addition of EMA records boosts the system recall to predict the suicidal ideation diagnosis from 48.13% obtained exclusively from EHR-based state-of-the-art methods to 67.78%. Additionally, our method provides interpretability through the t-SNE representation of the latent space. Further, the most relevant input features are identified and interpreted medically.

cs.LG↗

arXiv2019

On Generalized LDPC Codes for 5G Ultra Reliable Communication

Generalized low-density parity-check (GLDPC) codes, where single parity-check (SPC) constraint nodes are replaced with generalized constraint (GC) nodes, are a promising class of codes for low latency communication. In this paper, a practical construction of quasi-cyclic (QC) GLDPC codes is proposed, where the proportion of generalized constraints is determined by an asymptotic analysis. We analyze the message passing process and complexity of a GLDPC code over the additive white gaussian noise (AWGN) channel and present a constraint-to-variable update rule based on the specific codewords of the component code. The block error rate (BLER) performance of the GLDPC codes, combined with a complementary outer code, is shown to outperform a variety of state-of-the-art code and decoder designs with suitable lengths and rates for the 5G Ultra Reliable Communication (URC) regime over an additive white gaussian noise (AWGN) channel with quadrature PSK (QPSK) modulation.

cs.IT↗

arXiv2019

Probabilistic Time of Arrival Localization

In this paper, we take a new approach for time of arrival geo-localization. We show that the main sources of error in metropolitan areas are due to environmental imperfections that bias our solutions, and that we can rely on a probabilistic model to learn and compensate for them. The resulting localization error is validated using measurements from a live LTE cellular network to be less than 10 meters, representing an order-of-magnitude improvement.

cs.LG↗

arXiv2019

Continuous Transmission of Spatially-Coupled LDPC Code Chains

We propose a novel encoding/transmission scheme called continuous chain (CC) transmission that is able to improve the finite-length performance of a system using spatially-coupled low-density parity-check (SC-LDPC) codes. In CC transmission, instead of transmitting a sequence of independent codewords from a terminated SC-LDPC code chain, we connect multiple chains in a layered format, where encoding, transmission, and decoding are now performed in a continuous fashion. The connections between chains are created at specific points, chosen to improve the finite-length performance of the code structure under iterative decoding. We describe the design of CC schemes for different SC-LDPC code ensembles constructed from protographs: a (J,K)-regular SC-LDPC code chain, a spatially-coupled repeat-accumulate (SC-RA) code, and a spatially-coupled accumulate-repeat-jagged-accumulate (SC- ARJA) code. In all cases, significant performance improvements are reported and, in addition, it is shown that using CC transmission only requires a small increase in decoding complexity and decoding delay with respect to a system employing a single SC-LDPC code chain for transmission.

cs.IT↗

arXiv2019

Probabilistic MIMO Symbol Detection with Expectation Consistency Approximate Inference

In this paper we explore low-complexity probabilistic algorithms for soft symbol detection in high-dimensional multiple-input multiple-output (MIMO) systems. We present a novel algorithm based on the Expectation Consistency (EC) framework, which describes the approximate inference problem as an optimization over a non-convex function. EC generalizes algorithms such as Belief Propagation and Expectation Propagation. For the MIMO symbol detection problem, we discuss feasible methods to find stationary points of the EC function and explore their tradeoffs between accuracy and speed of convergence. The accuracy is studied, first in terms of input-output mutual information and show that the proposed EC MIMO detector greatly improves state-of-the-art methods, with a complexity order cubic in the number of transmitting antennas. Second, these gains are corroborated by combining the probabilistic output of the EC detector with a low-density parity-check (LDPC) channel code.

cs.IT↗

arXiv2019

Out-of-Sample Testing for GANs

We propose a new method to evaluate GANs, namely EvalGAN. EvalGAN relies on a test set to directly measure the reconstruction quality in the original sample space (no auxiliary networks are necessary), and it also computes the (log)likelihood for the reconstructed samples in the test set. Further, EvalGAN is agnostic to the GAN algorithm and the dataset. We decided to test it on three state-of-the-art GANs over the well-known CIFAR-10 and CelebA datasets.

cs.LG↗