Humanitarian applications of machine learning with remote-sensing data: review and case study in refugee settlement mapping.

The coordination of humanitarian relief, e.g. in a natural disaster or a conflict situation, is often complicated by a scarcity of data to inform planning. Remote sensing imagery, from satellites or drones, can give important insights into conditions on the ground, including in areas which are difficult to access. Applications include situation awareness after natural disasters, structural damage assessment in conflict, monitoring human rights violations or population estimation in settlements. We review machine learning approaches for automating these problems, and discuss their potential and limitations. We also provide a case study of experiments using deep learning methods to count the numbers of structures in multiple refugee settlements in Africa and the Middle East. We find that while high levels of accuracy are possible, there is considerable variation in the characteristics of imagery collected from different sensors and regions. In this, as in the other applications discussed in the paper, critical inferences must be made from a relatively small amount of pixel data. We, therefore, consider that using machine learning systems as an augmentation of human analysts is a reasonable strategy to transition from current fully manual operational pipelines to ones which are both more efficient and have the necessary levels of quality control.This article is part of a discussion meeting issue 'The growing ubiquity of algorithms in society: implications, impacts and innovations'.

Direct learning of sparse changes in Markov networks by density ratio estimation.

We propose a new method for detecting changes in Markov network structure between two sets of samples. Instead of naively fitting two Markov network models separately to the two data sets and figuring out their difference, we directly learn the network structure change by estimating the ratio of Markov network models. This density-ratio formulation naturally allows us to introduce sparsity in the network structure change, which highly contributes to enhancing interpretability. Furthermore, computation of the normalization term, a critical bottleneck of the naive approach, can be remarkably mitigated. We also give the dual formulation of the optimization problem, which further reduces the computation cost for large-scale Markov networks. Through experiments, we demonstrate the usefulness of our method.

Factorial switching linear dynamical systems applied to physiological condition monitoring.

Condition monitoring often involves the analysis of systems with hidden factors that switch between different modes of operation in some way. Given a sequence of observations, the task is to infer the filtering distribution of the switch setting at each time step. In this paper, we present factorial switching linear dynamical systems as a general framework for handling such problems. We show how domain knowledge and learning can be successfully combined in this framework, and introduce a new factor (the "X-factor") for dealing with unmodeled variation. We demonstrate the flexibility of this type of model by applying it to the problem of monitoring the condition of a premature baby receiving intensive care. The state of health of a baby cannot be observed directly, but different underlying factors are associated with particular patterns of physiological measurements and artifacts. We have explicit knowledge of common factors and use the X-factor to model novel patterns which are clinically significant but have unknown cause. Experimental results are given which show the developed methods to be effective on typical intensive care unit monitoring data.