Inferring Nighttime Satellite Imagery from Human Mobility
Authors: Brian Dickinson, Gourab Ghoshal, Xerxes Dotiwalla, Adam Sadilek, Henry Kautz394-402
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this study we demonstrate that it is possible to accelerate this process by inferring artificial nighttime satellite imagery from human mobility data, while maintaining a strong differential privacy guarantee. We also show that these artificial maps can be used to infer socioeconomic variables, often with greater accuracy than using actual satellite imagery. Along the way, we find that the relationship between mobility and light emissions is both nonlinear and varies considerably around the globe. Finally, we show that models based on human mobility can significantly improve our understanding of society at a global scale. |
| Researcher Affiliation | Collaboration | Brian Dickinson,1 Gourab Ghoshal,1 Xerxes Dotiwalla,2 Adam Sadilek,2 Henry Kautz1 1University of Rochester 500 Joseph C. Wilson Blvd. Rochester, New York 14627 2Google Inc. 1600 Amphitheatre Parkway Mountain View, California 94043 |
| Pseudocode | No | The paper provides a high-level outline of the method steps in bullet points, but it does not include a structured pseudocode block or an explicitly labeled algorithm. |
| Open Source Code | No | The paper does not provide an explicit statement or link indicating that the source code for the described methodology is publicly available. It mentions using Google's Location History Data. |
| Open Datasets | No | The paper states: "We propose to use anonymous and aggregated flows from users opted-in to Google s Location History a global data source..." and "The Google Mobility dataset is a heavily aggregated and anonymized summary of global trips mostly provided by location services on Android phones of users who have enabled location history." This is a proprietary dataset from Google, and no public access information (link, repository) is provided. |
| Dataset Splits | Yes | Using 5-fold cross-fold validation, performing linear regression on our 2016 annual dataset resulted in prediction a mean absolute error of 178.47." and "To combat this, we make two significant changes to our cross-fold validation procedure. First we segment all of our data into over 50, 000 1 longitude by 1 longitude blocks, of which about 7, 000 actually have some mobility information. Learning models for each of these blocks separately significantly limits the proximity of cells in the training and testing sets. Our second change additionally excludes blocks adjacent to predicted block from the training set. |
| Hardware Specification | No | The paper does not specify the hardware used for running the experiments. It only mentions "several weeks of compute time" without details. |
| Software Dependencies | No | The paper describes the use of statistical models like linear regression and random forest, and techniques like kriging, but does not provide specific software dependencies with version numbers (e.g., Python 3.x, scikit-learn x.x.x, PyTorch x.x.x). |
| Experiment Setup | No | The paper describes the types of models used (linear regression, random forest), the features extracted from mobility data, and the general approach (kriging vs. regional models, cross-validation setup), but it does not provide specific experimental setup details such as learning rates, batch sizes, number of epochs, or optimizer settings. |