Two Surgo papers were accepted into NeurIPS2019 Workshops. For the workshop on“Do the right thing”: machine learning and causal inference for improved decision making, our paper “Causal datasheet: An approximate guide to practically assess Bayesian networks in the real world” discusses a new tool for users to validate results and define uncertainty for Bayesian Network analysis for any dataset. In the AI for Social Good workshop, we will present “ML and precision public health: Saving mothers and babies from dying in rural India", which demonstrates how optimized data collection coupled with machine learning provides a holistic picture into the drivers and barriers to hospital delivery.
In both health and education, we see that interventions focusing in a single area (teachers, nurses, district officials, etc.) often do not have their intended impact – Why? Regardless of whether we are talking about improving teacher practices in the classroom or getting nurses to consistently check the blood pressure of mothers in labor, the answer to this question often includes the lack of system alignment and motivation for change. Surgo and STiR convened stakeholders across the global development sector for a candid discussion around how to harness intrinsic motivation for system change across global development sectors.