• We build a contextualized Twitter-based language model to construct poverty metrics. • Twitter discourse predicts village-level poverty. • Less affluent communities focus more on concrete, local development concerns. • Kriging interpolation improves sparse social media data with uncertainty estimates. • We propose a novel data pipeline for poverty assessment which utilizes citizen participation. We present a novel pipeline for poverty assessment using social media and apply it to a dataset of 1.2 million geotagged tweets from Zambia. Leveraging mixed-methods topic modeling with domain-guided feature selection, we develop an interpretable language model that explains more than 60 % of the variation in village-level wealth. Our findings show that the tweets from poorer villages emphasize local, concrete needs, whereas those from wealthier villages focus on abstract development concepts. We also compare imputation methods for data-sparse contexts and find that kriging improves predictive accuracy by 15 % over standard approaches, while providing uncertainty quantification for adaptive sampling. This work demonstrates the viability of social media discourse as a participatory, scalable poverty monitoring tool in regions with limited data.

This paper proposes an integration of surrogate modeling and topology to significantly reduce the amount of data required to describe the underlying global dynamics of robot controllers, including closed-box ones. A Gaussian Process (GP), trained with randomized short trajectories over the state-space, acts as a surrogate model for the underlying dynamical system. Then, a combinatorial representation is built and used to describe the dynamics in the form of a directed acyclic graph, known as Morse graph. The Morse graph is able to describe the system's attractors and their corresponding regions of attraction (RoA). Furthermore, a pointwise confidence level of the global dynamics estimation over the entire state space is provided. In contrast to alternatives, the framework does not require estimation of Lyapunov functions, alleviating the need for high prediction accuracy of the GP. The framework is suit-able for data-driven controllers that do not expose an analytical model as long as Lipschitz-continuity is satisfied. The method is compared against established analytical and recent machine learning alternatives for estimating Roas, outperforming them in data efficiency without sacrificing accuracy. Link to code: https://go.rutgers.edu/49hy35en

Motivated by the problem of optimization of force-field systems in physics using large-scale computer simulations, we consider exploration of a deterministic complex multivariate response surface. The objective is to find input combinations that generate output close to some desired or “target” vector. Despite reducing the problem to exploration of the input space with respect to a 1-D loss function, the search is nontrivial and challenging due to infeasible input combinations, high dimensionalities of the input and output space and multiple “desirable” regions in the input space, and the difficulty of emulating the objective function well with a surrogate model. We propose an approach that is based on combining machine learning techniques with smart experimental design ideas to locate multiple good regions in the input space. Note to Practitioners—ReaxFF is a force field that incorporates complex functions with associated inputs in order to describe the inter- and intra-atomic interactions in materials systems. A typical ReaxFF force field consists of hundreds of parameters (inputs) per element type. During the development of a force field for a molecular system of interest, using computer simulations, these parameters are optimized to reproduce hundreds of material properties close to some benchmark reference values. Finding “good” combinations of hundreds of parameters that produce hundreds of reference values close to their gold standards is a challenging problem because there may be several parameter combinations that may be “almost equally good” or “equally desirable.” To add to the complication, several input combinations simply lead to a system crash, not producing any output at all. Standard global optimization methods do not address such a problem. We propose a novel framework that can address this problem. Beyond the ReaxFF optimization, it can be applied to multiobjective optimization in engineering and the physical sciences, where there are unknown constraints and the focus is on obtaining several good points that can serve as alternatives to a single global optimum.

We propose a modified variational autoencoder (VAE) architecture built on top of Mask R-CNN for instance-level video segmentation and tracking. The method builds a shared encoder and three parallel decoders, yielding three disjoint branches for predictions of future frames, object detection boxes, and instance segmentation masks. To effectively solve multiple learning tasks, we introduce a Gaussian Process model to enhance the statistical representation of VAE by relaxing the prior strong independent and identically distributed (iid) assumption of conventional VAEs and allowing potential correlations among extracted latent variables. The network learns embedded spatial interdependence and motion continuity in video data and creates a representation that is effective to produce high-quality segmentation masks and track multiple instances in diverse and unstructured videos. Evaluation on a variety of recently introduced datasets shows that our model outperforms previous methods and achieves the new best in class performance.

This paper presents a prior-less method for tracking and clustering an unknown number of human faces and maintaining their individual identities in unconstrained videos. The key challenge is to accurately track faces with partial occlusion and drastic appearance changes in multiple shots resulting from significant variations of makeup, facial expression, head pose and illumination. To address this challenge, we propose a new multi-face tracking and re-identification algorithm, which provides high accuracy in face association in the entire video with automatic cluster number generation, and is robust to outliers. We develop a co-occurrence model of multiple body parts to seamlessly create face tracklets, and recursively link tracklets to construct a graph for extracting clusters. A Gaussian Process model is introduced to compensate the deep feature insufficiency, and is further used to refine the linking results. The advantages of the proposed algorithm are demonstrated using a variety of challenging music videos and newly introduced body-worn camera videos. The proposed method obtains significant improvements over the state of the art [51], while relying less on handling video-specific prior information to achieve high performance.

MOTIVATION: Abrupt reduction/resumption of thermal fluctuations of a force probe has been used to identify association/dissociation events of protein-ligand bonds. We show that off-rate of molecular dissociation can be estimated by the analysis of the bond lifetime, while the on-rate of molecular association can be estimated by the analysis of the waiting time between two neighboring bond events. However, the analysis relies heavily on subjective judgments and is time-consuming. To automate the process of mapping out bond events from thermal fluctuation data, we develop a hidden Markov model (HMM)-based method. RESULTS: The HMM method represents the bond state by a hidden variable with two values: bound and unbound. The bond association/dissociation is visualized and pinpointed. We apply the method to analyze a key receptor-ligand interaction in the early stage of hemostasis and thrombosis: the von Willebrand factor (VWF) binding to platelet glycoprotein Ibα (GPIbα). The numbers of bond lifetime and waiting time events estimated by the HMM are much more than those estimated by a descriptive statistical method from the same set of raw data. The kinetic parameters estimated by the HMM are in excellent agreement with those by a descriptive statistical analysis, but have much smaller errors for both wild-type and two mutant VWF-A1 domains. Thus, the computerized analysis allows us to speed up the analysis and improve the quality of estimates of receptor-ligand binding kinetics.

Abstract This paper describes the design, implementation, and performance of ES‐Kit, a distributed object‐oriented system being developed by the Experimental Systems Project at the Microelectronics and Computer Technology Corporation. The operating system consists of a kernel and a set of Public Service Objects which dynamically extend the functionality of the kernel by providing several traditional operating system services when required by application objects. Applications for the ES‐Kit environment are written in GNU C++ and do not require additional language primitives for distributed execution. Initial performance results from a representative set of applications indicate that the object‐oriented paradigm provides a powerful solution to distributed programming.