Emily Aiken

CV | Google Scholar | GitHub | emilyaiken@berkeley.edu


I am a fifth year PhD student at the UC Berkeley School of Information advised by Joshua Blumenstock. My research interests are in machine learning and development economics, with a focus on analyzing large digital traces to inform social protection policy. Before Berkeley, I received my BA in computer science from Harvard University where I conducted research on digital epidemiology with Mauricio Santillana. I am funded by a Microsoft Research PhD Fellowship and am affiliated with the Global Policy Lab and Berkeley AI Research (BAIR).

★ I am on the academic job market (for tenure track positions and postdocs) in 2023-2024!

Recent News (See all news)

Three new working papers out, on (1) estimating cash impacts with digital data, (2) constructing differentially private mobility matrices, and (3) updating proxy-means tests (extended abstract only, full working paper to come).
I'm attending COMPASS '23! Our paper on proxy means test updating is in Session 6 (4:50pm on Thursday) and I'll be presenting my dissertation research in the doctoral consortium.
I will give a talk on impact evaluation from digital data at the HBS Workshop on the Future of AI and Economics on July 31.
I am co-organizing a session on digital data at the World Bank/GIZ Global Forum on Adaptive Social Protection.
I will be at ICLR '23. Find our work presented at the workshop on Practical ML for Development (10:15am on Friday) and the workshop on ML for Remote Sensing (3pm on Friday).

Publications and Working Papers

Journal Articles and Conference Proceedings

Fairness and representation in satellite-based poverty maps:
Evidence of urban-rural disparities and their impacts on downstream policy.
E. Aiken*, E. Rolf*, and J. Blumenstock.
Accepted, IJCAI (2023).
Machine learning and phone data can improve targeting of humanitarian aid.
E. Aiken, S. Bellue, D. Karlan, C. Udry, and J. Blumenstock.
Nature 603, No. 7903 (2022).
★ Cover article
Early versions: NBER Working Paper No. 29070 (2022).
Phone sharing and cash transfers in Togo: Quantitative evidence from mobile phone data.
E. Aiken*, V. Thakur*, and J. Blumenstock.
★ Best paper award
Targeting development aid with machine learning and mobile phone data:
Evidence from an anti-poverty intervention in Afghanistan.
E. Aiken, G. Bedoya, A. Coville, and J. Blumenstock.
Journal of Development Economics 161 (2022).
Early verisons: World Bank Policy Research Working Paper No. 10252 (2022). Extended abstract at ACM COMPASS (2020).
Towards the use of neural networks for influenza prediction at multiple spatial resolutions.
E. Aiken, A. Nguyen, C. Viboud, and M. Santillana.
Science Advances 7, No. 8 (2021).
Early versions: Extended abstract at NeurIPS ML for Health (2019).
Real-time estimation of disease activity in emerging outbreaks using internet search information.
E. Aiken, S. McGough, M. Majumder, G. Wachtel, A. Nguyen, C. Viboud, and M. Santillana.
PLoS Computational Biology 16, No. 8 (2020).


Preprints and Working Papers

Estimating impacts with surveys vs. digital traces: Evidence from randomized cash transfers in Togo.
E. Aiken, S. Bellue, D. Karlan, C. Udry, and J. Blumenstock.
Working Paper (2023).
Privacy gaurantees for personal mobility data in humanitarian response.
N. Kohli*, E. Aiken*, and J. Blumenstock.
arXiv Preprint (2023).
Early versions: KDD Workshop on Humanitarian Mapping (2020).


Workshop and Short Papers

Moving targets: When does a poverty prediction model need to be updated?
E. Aiken*, T. Ohlenburg*, and J. Blumenstock.
ACM COMPASS, Short Papers Track (2023).
Can strategic data collection improve the performance of poverty prediction models?
S. Soman, E. Aiken, E. Rolf, and J. Blumenstock.
ICLR Workshop on Practical ML for Development (2023).
Home location detection from mobile phone data: Evidence from Togo.
R. Warren, E. Aiken, and J. Blumenstock.
ACM COMPASS, Poster Track (2022).

*Equal Contribution