The Sustainable Development Goals (SDGs) of the United Nations provide a blueprint of a better future by 'leaving no one behind', and, to achieve the SDGs by 2030, poor countries require immense volumes of development aid. In this paper, we develop a causal machine learning framework for predicting heterogeneous treatment effects of aid disbursements to inform effective aid allocation. Specifically, our framework comprises three components: (i) a balancing autoencoder that uses representation learning to embed high-dimensional country characteristics while addressing treatment selection bias; (ii) a counterfactual generator to compute counterfactual outcomes for varying aid volumes to address small sample-size settings; and (iii) an inference model that is used to predict heterogeneous treatment-response curves. We demonstrate the effectiveness of our framework using data with official development aid earmarked to end HIV/AIDS in 105 countries, amounting to more than USD 5.2 billion. For this, we first show that our framework successfully computes heterogeneous treatment-response curves using semi-synthetic data. Then, we demonstrate our framework using real-world HIV data. Our framework points to large opportunities for a more effective aid allocation, suggesting that the total number of new HIV infections could be reduced by up to 3.3% (~50,000 cases) compared to the current allocation practice.

Originally posted on The Horizons Tracker. As William Easterly explains in The White Man's Burden, while huge sums have been spent on aid to developing countries, much has been wasted due to the supposed hubris of the west. Easterly argues that much of development work is the preserve of so-called "planners", who concoct investment schemes from their offices in western cities and then impose them on the recipient community. Research1 from ETH Zurich ponders whether AI might be able to do a better job. The researchers suggest that it is often difficult to get an accurate overview of the work being done due to the multitude of projects and institutions supporting them.

It sounds like a dream: Poor villagers are handed money regularly every month, for several years, with no conditions attached. An American organization is currently testing the model in Kenya. When the village elder came to her in September to invite her to a meeting under the acacia trees, Norah Odhiambo was skeptical. Storm clouds were gathering above nearby Lake Victoria, the 34-year-old relates, and she set aside the machete she uses to clear brush from her neighbor's field for a few shillings a day. A new aid organization called GiveDirectly, the village elder said, would like to introduce itself.

When a technology has its breakthrough, can often only be determined in hindsight. In the case of artificial intelligence (AI) and machine learning (ML), this is different. ML is that part of AI that describes rules and recognizes patterns from large amounts of data in order to predict future data. Both concepts are virtually omnipresent and at the top of most buzzword rankings. Personally, I think – and this is clearly linked to the rise of AI and ML – that there has never been a better time than today to develop smart applications and use them.