Sang Michael Xie

Reliable data on economic livelihoods remain scarce in the developing world, hampering efforts to study these outcomes and to design policies that improve them. Here we demonstrate an accurate, inexpensive, and scalable method for estimating consumption expenditure and asset wealth from high-resolution satellite imagery. Using survey and satellite data from five African countries--Nigeria, Tanzania, Uganda, Malawi, and Rwanda--we show how a convolutional neural network can be trained to identify image features that can explain up to 75% of the variation in local-level economic outcomes. Our method, which requires only publicly available data, could transform efforts to track and target poverty in developing countries. It also demonstrates how powerful machine learning techniques can be applied in a setting with limited training data, suggesting broad potential application across many scientific domains.

The lack of reliable data in developing countries is a major obstacle to sustainable development, food security, and disaster relief. Poverty data, for example, is typically scarce, sparse in coverage, and labor-intensive to obtain. Remote sensing data such as high-resolution satellite imagery, on the other hand, is becoming increasingly available and inexpensive. Unfortunately, such data is highly unstructured and currently no techniques exist to automatically extract useful insights to inform policy decisions and help direct humanitarian efforts. We propose a novel machine learning approach to extract large-scale socioeconomic indicators from high-resolution satellite imagery. The main challenge is that training data is very scarce, making it difficult to apply modern techniques such as Convolutional Neural Networks (CNN). We therefore propose a transfer learning approach where nighttime light intensities are used as a data-rich proxy. We train a fully convolutional CNN model to predict nighttime lights from daytime imagery, simultaneously learning features that are useful for poverty prediction. The model learns filters identifying different terrains and man-made structures, including roads, buildings, and farmlands, without any supervision beyond nighttime lights. We demonstrate that these learned features are highly informative for poverty mapping, even approaching the predictive performance of survey data collected in the field.

Distribution shifts -- where the training distribution differs from the test distribution -- can substantially degrade the accuracy of machine learning (ML) systems deployed in the wild. Despite their ubiquity in the real-world deployments, these distribution shifts are under-represented in the datasets widely used in the ML community today. To address this gap, we present WILDS, a curated benchmark of 10 datasets reflecting a diverse range of distribution shifts that naturally arise in real-world applications, such as shifts across hospitals for tumor identification; across camera traps for wildlife monitoring; and across time and location in satellite imaging and poverty mapping. On each dataset, we show that standard training yields substantially lower out-of-distribution than in-distribution performance. This gap remains even with models trained by existing methods for tackling distribution shifts, underscoring the need for new methods for training models that are more robust to the types of distribution shifts that arise in practice. To facilitate method development, we provide an open-source package that automates dataset loading, contains default model architectures and hyperparameters, and standardizes evaluations. Code and leaderboards are available at https://wilds.stanford.edu.

Accurate automated segmentation of remote sensing data could benefit applications from land cover mapping and agricultural monitoring to urban development surveyal and disaster damage assessment. While convolutional neural networks (CNNs) achieve state-of-the-art accuracy when segmenting natural images with huge labeled datasets, their successful translation to remote sensing tasks has been limited by low quantities of ground truth labels, especially fully segmented ones, in the remote sensing domain. In this work, we perform cropland segmentation using two types of labels commonly found in remote sensing datasets that can be considered sources of “weak supervision”: (1) labels comprised of single geotagged points and (2) image-level labels. We demonstrate that (1) a U-Net trained on a single labeled pixel per image and (2) a U-Net image classifier transferred to segmentation can outperform pixel-level algorithms such as logistic regression, support vector machine, and random forest. While the high performance of neural networks is well-established for large datasets, our experiments indicate that U-Nets trained on weak labels outperform baseline methods with as few as 100 labels. Neural networks, therefore, can combine superior classification performance with efficient label usage, and allow pixel-level labels to be obtained from image labels.

While adversarial training can improve robust accuracy (against an adversary), it sometimes hurts standard accuracy (when there is no adversary). Previous work has studied this tradeoff between standard and robust accuracy, but only in the setting where no predictor performs well on both objectives in the infinite data limit. In this paper, we show that even when the optimal predictor with infinite data performs well on both objectives, a tradeoff can still manifest itself with finite data. Furthermore, since our construction is based on a convex learning problem, we rule out optimization concerns, thus laying bare a fundamental tension between robustness and generalization. Finally, we show that robust self-training mostly eliminates this tradeoff by leveraging unlabeled data.