HVGH: Unsupervised Segmentation for High-Dimensional Time Series Using Deep Neural Compression and Statistical Generative Model

Spatio-temporal categorization for first-person-view videos using a convolutional variational autoencoder and Gaussian processes

In this study, HcVGH, a method that learns spatio-temporal categories by segmenting first-person-view (FPV) videos captured by mobile robots, is proposed. Humans perceive continuous high-dimensional information by dividing and categorizing it into significant segments. This unsupervised segmentation capability is considered important for mobile robots to learn spatial knowledge. The proposed HcVGH combines a convolutional variational autoencoder (cVAE) with HVGH, a past method, which follows the hierarchical Dirichlet process-variational autoencoder-Gaussian process-hidden semi-Markov model comprising deep generative and statistical models. In the experiment, FPV videos of an agent were used in a simulated maze environment. FPV videos contain spatial information, and spatial knowledge can be learned by segmenting them. Using the FPV-video dataset, the segmentation performance of the proposed model was compared with previous models: HVGH and hierarchical recurrent state space model. The average segmentation F-measure achieved by HcVGH was 0.77; therefore, HcVGH outperformed the baseline methods. Furthermore, the experimental results showed that the parameters that represent the movability of the maze environment can be learned.

Identifying the Socioeconomic, Demographic, and Political Determinants of Social Mobility and Their Effects on COVID-19 Cases and Deaths: Evidence From US Counties

Background

The spread of COVID-19 at the local level is significantly impacted by population mobility. The U.S. has had extremely high per capita COVID-19 case and death rates. Efficient nonpharmaceutical interventions to control the spread of COVID-19 depend on our understanding of the determinants of public mobility.

Objective

This study used publicly available Google data and machine learning to investigate population mobility across a sample of US counties. Statistical analysis was used to examine the socioeconomic, demographic, and political determinants of mobility and the corresponding patterns of per capita COVID-19 case and death rates.

Methods

Daily Google population mobility data for 1085 US counties from March 1 to December 31, 2020, were clustered based on differences in mobility patterns using K-means clustering methods. Social mobility indicators (retail, grocery and pharmacy, workplace, and residence) were compared across clusters. Statistical differences in socioeconomic, demographic, and political variables between clusters were explored to identify determinants of mobility. Clusters were matched with daily per capita COVID-19 cases and deaths.

Results

Our results grouped US counties into 4 Google mobility clusters. Clusters with more population mobility had a higher percentage of the population aged 65 years and over, a greater population share of Whites with less than high school and college education, a larger percentage of the population with less than a college education, a lower percentage of the population using public transit to work, and a smaller share of voters who voted for Clinton during the 2016 presidential election. Furthermore, clusters with greater population mobility experienced a sharp increase in per capita COVID-19 case and death rates from November to December 2020.

Conclusions

Republican-leaning counties that are characterized by certain demographic characteristics had higher increases in social mobility and ultimately experienced a more significant incidence of COVID-19 during the latter part of 2020.

Robot Concept Acquisition Based on Interaction Between Probabilistic and Deep Generative Models

We propose a method for multimodal concept formation. In this method, unsupervised multimodal clustering and cross-modal inference, as well as unsupervised representation learning, can be performed by integrating the multimodal latent Dirichlet allocation (MLDA)-based concept formation and variational autoencoder (VAE)-based feature extraction. Multimodal clustering, representation learning, and cross-modal inference are critical for robots to form multimodal concepts from sensory data. Various models have been proposed for concept formation. However, in previous studies, features were extracted using manually designed or pre-trained feature extractors and representation learning was not performed simultaneously. Moreover, the generative probabilities of the features extracted from the sensory data could be predicted, but the sensory data could not be predicted in the cross-modal inference. Therefore, a method that can perform clustering, feature learning, and cross-modal inference among multimodal sensory data is required for concept formation. To realize such a method, we extend the VAE to the multinomial VAE (MNVAE), the latent variables of which follow a multinomial distribution, and construct a model that integrates the MNVAE and MLDA. In the experiments, the multimodal information of the images and words acquired by a robot was classified using the integrated model. The results demonstrated that the integrated model can classify the multimodal information as accurately as the previous model despite the feature extractor learning in an unsupervised manner, suitable image features for clustering can be learned, and cross-modal inference from the words to images is possible.