This study quantifies the impact of 6,068 non-pharmaceutical interventions (NPIs) implemented in 79 territories on the effective reproduction number, $R_e$, of COVID-19. The authors propose a modeling approach that combines four computational techniques—case-control analysis (CC), LASSO time-series regression, random forests (RF), and transformers (TF)—to assess the effectiveness of NPIs. They validate their findings using two external datasets covering 42,151 additional NPIs from 226 countries. The results indicate that a combination of less disruptive and costly NPIs can be as effective as more intrusive measures, such as national lockdowns. The study identifies six NPI categories that show significant impacts on $R_e$: small gathering cancellations, closure of educational institutions, border restrictions, increased availability of personal protective equipment (PPE), individual movement restrictions, and national lockdowns. The effectiveness of NPIs varies across different territories, influenced by factors such as timing of implementation, local context, and governance indicators. The study also explores the importance of risk-communication strategies, government assistance to vulnerable populations, and the timing of NPI adoption. The findings suggest that no single NPI can effectively reduce $R_e$ below one, but a suitable combination of NPIs tailored to the specific country and its epidemic stage can significantly curb the spread of COVID-19.This study quantifies the impact of 6,068 non-pharmaceutical interventions (NPIs) implemented in 79 territories on the effective reproduction number, $R_e$, of COVID-19. The authors propose a modeling approach that combines four computational techniques—case-control analysis (CC), LASSO time-series regression, random forests (RF), and transformers (TF)—to assess the effectiveness of NPIs. They validate their findings using two external datasets covering 42,151 additional NPIs from 226 countries. The results indicate that a combination of less disruptive and costly NPIs can be as effective as more intrusive measures, such as national lockdowns. The study identifies six NPI categories that show significant impacts on $R_e$: small gathering cancellations, closure of educational institutions, border restrictions, increased availability of personal protective equipment (PPE), individual movement restrictions, and national lockdowns. The effectiveness of NPIs varies across different territories, influenced by factors such as timing of implementation, local context, and governance indicators. The study also explores the importance of risk-communication strategies, government assistance to vulnerable populations, and the timing of NPI adoption. The findings suggest that no single NPI can effectively reduce $R_e$ below one, but a suitable combination of NPIs tailored to the specific country and its epidemic stage can significantly curb the spread of COVID-19.