View GitHub Repository responsibly

Schedule Inspection
Subscribe

responsibly.fairness.interventions   A
last analyzed

↳ Parent: Project

Complexity

Total Complexity 0

Size/Duplication

Total Lines 2
Duplicated Lines 0 %

Importance

Changes 0
Metric Value
eloc 1
dl 0
loc 2
rs 10
c 0
b 0
f 0
wmc 0
1 
"""
2 
Algorithmic intervensions for satisfying fairness criteria.
3 

            


            
                                                                                                            
                            
            
                                    
            
            

                
                    4
                

                                    
                                                    
                
                
                
There are three families of techniques:

            


            
                                                                                                            
                            
            
                                    
            
            

                
                    5
                

                                    
                                                    
                
                
                

            


            
                                                                                                            
                            
            
                                    
            
            

                
                    6
                

                                    
                                                    
                
                
                
1. **Pre-processing** - Adjust features in the dataset.

            


            
                                                                                                            
                            
            
                                    
            
            

                
                    7
                

                                    
                                                    
                
                
                
2. **In-processing** - Adjust the learning algorithm.

            


            
                                                                                                            
                            
            
                                    
            
            

                
                    8
                

                                    
                                                    
                
                
                
3. **Post-processing** - Adjust the learned classifier.

            


            
                                                                                                            
                            
            
                                    
            
            

                
                    9
                

                                    
                                                    
                
                
                

            


            
                                                                                                            
                                                                
            
                                    
            
            

                
                    10
                

                                    
                                                    
                
                
                
"""

            


            
                                                        
            
                                    
            
            

                
                    11