update readme

Author: jmejia8

README.md

@@ -1,6 +1,6 @@
-# Euclidean Distance Project
+# Infeasibility detector for Multiobjective Bilevel Optimization
 
-This project computes the Euclidean distance between all rows of a dataset.
+This project implements the labeling process to identify infeasible solutions in multi-objective bilevel optimization.
 
 ## Project Structure
 
@@ -19,8 +19,26 @@ This project computes the Euclidean distance between all rows of a dataset.
    make
    ```
 
-2. Run the program:
+2. Run the program to obtain the labels for an archive, with `epsilon=0.001` (`1e-8` by default if not provided) and known Pareto-optimal front (optional):
 
    ```
-   ./bin/main data.csv 0.001
+   ./bin/detector archive.csv 0.001 pof.csv
    ```
+
+The `archive.csv` is a comma-separated value file, which must contain the decision variables, UL objectives and LL objectives by row. The program assumes that solutions violating equality and inequality constraints are not included. The following table illustrates the necessary format.
+
+```csv
+x1,x2,F1,F2,f1,f2
+0.58,0.42,0.16,0.52,0.75,0.84
+0.59,0.14,0.51,0.45,0.81,0.86
+0.52,0.94,0.06,0.63,0.58,0.79
+0.37,0.32,0.92,0.45,0.87,0.84
+0.62,0.73,0.39,0.44,0.09,0.47
+0.90,0.90,0.37,0.81,0.23,0.78
+0.34,0.03,0.88,0.06,0.01,0.44
+```
+
+The header specifies each upper-level decisions **x**, upper-level objectives **F**, lower-level objectives **f**.
+A similar format is required for the Pareto-optimal front `pof.csv`, only upper-level objectives are mandatory.
+
+3. The resulting labels are saved next to `archive.csv` with name `archive.csv_labels.csv`