Added ref to the arXiv paper.

Author: dmachlanski

README.md

@@ -1,6 +1,6 @@
 # Undersmoothing Causal Estimators with Generative Trees
 
-This code accompanies the paper *Undersmoothing Causal Estimators with Generative Trees*.
+This code accompanies the paper *Undersmoothing Causal Estimators with Generative Trees* [1].
 
 The experimental setup is based on the [CATE benchmark](https://github.com/misoc-mml/cate-benchmark) and extended to test our proposed DeGeTs framework.
 
@@ -31,4 +31,7 @@ experiments/demo.sh - A small experimental run for demonstration purposes. It in
 Note the intermediate results printed to the console while running the scripts show mean +- standard error. The final results in the paper were computed separately from outputted CSV files in order to show mean +- 95% confidence intervals.
 
 ## More Info
-For more details about how to use the framework and analyse the results, please visit the [CATE benchmark](https://github.com/misoc-mml/cate-benchmark) website.
+For more details about how to use the framework and analyse the results, please visit the [CATE benchmark](https://github.com/misoc-mml/cate-benchmark) website.
+
+## References
+[1] D. Machlanski, S. Samothrakis, and P. Clarke, ‘Undersmoothing Causal Estimators with Generative Trees’, arXiv:2203.08570 [cs, stat], Mar. 2022, doi: 10.48550/arXiv.2203.08570.