Tutorial 2: Fairness in Rankings and Recommenders
Table of Contents (powered by https://videoken.com)
0:01:45 [Tutorial 2: Fairness in Rankings and Recommenders]
0:03:53 Presenters
0:04:13 Algorithmic fairness: why?
0:04:52 Case Studies: Image Search
0:05:58 Case Studies: COMPAS
0:06:33 And many more
0:07:14 What is the cause: Data
0:07:42 What is the cause: Algorithms
0:08:00 Tutorial outline
0:08:29 Definition
0:09:06 Individual fairness
0:10:37 Group Fairness
0:11:10 Group Fairness in classification
0:11:46 Blindness is not enough
0:12:06 Disparate treatment vs disparate impact
0:12:51 Non-discrimination and equality of opportunity
0:13:29 Group fairness: base rates
0:14:57 Group fairness: criticism
0:15:42 Other definitions of fairness
0:17:01 Fairness in Ranking
0:18:40 Fairness constraints
0:19:17 Discounted cumulative fairness
0:21:50 Fairness of exposure
0:24:09 Equity of attention [BGW18]
0:25:09 Equity of amortized attention [BGW18]
0:26:05 Definition of fairness in ranking (summary)
0:26:44 Achieving fairness
0:27:24 Pre-processing
0:27:58 In processing: Learning to rank algorithms
0:30:11 In processing: learning fair representations
0:33:19 Post-processing
0:33:35 Post-processing: generative process
0:36:09 Post processing: Constraint optimization problem
0:36:54 Post-processing: LP optimization [SJ18]
0:36:59 Post processing: Constraint optimization (ranking maximize
0:37:18 Ensuring fairness in ranking (summary)
0:38:03 Fair vs diverse rankings
0:38:31 HICOT
0:41:30 Fairness in Recommender's
0:41:38 Multi-sided Fairness
0:42:23 Multi-sided Fairness in Recommender's
0:43:54 Ensuring Fairness in Recommender's
0:44:10 Pre-processing Methods
0:44:57 In-processing Methods
0:45:22 The STEM Example
0:46:47 USE MF & Count Fairness
0:48:32 The Regularization Approach
0:50:22 Randomness in VAE Recommender's
0:51:12 Post-processing Methods
0:51:18 Calibration Method
0:51:39 Post-Processing Methods:
Fairness in Group Recommender's
0:51:49 Fairness in Group Recommendations
0:52:57 Individual Utility, Social Welfare & Fairness
0:54:05 Social Welfare & Fairness
0:54:33 Ensuring Fairness
0:54:57 Fairness via Pareto
0:56:35 m-Proportionality
0:57:24 m-Envy-Freeness
0:59:02 Fairness in Sequential Recommendations
1:00:05 Satisfaction & Disagreements
1:01:46 References
1:02:19 Program Fairness
1:02:47 Program Fairness Verification
1:05:28 Fairness-Aware Programming
1:07:57 Fairness: Beyond Ranking and Recommender's
1:08:13 Some examples
1:08:23 Fairness in resource allocation
1:09:03 Example: ROBUS
1:09:21 Conclusions
1:12:08 "Fair" has many meanings