Boosting Demographic Fairness of Face Attribute Classifiers via Latent Adversarial Representations.
Huimin ZengZhenrui YueLanyu ShangYang ZhangDong WangPublished in: Big Data (2022)
Keyphrases
- adaboost algorithm
- weak classifiers
- ensemble learning
- weak learners
- randomized trees
- ensemble classifier
- haar like features
- boosting algorithms
- improving classification accuracy
- boosting framework
- boosted classifiers
- feature selection
- face detector
- strong classifier
- accurate classifiers
- multiple classifier systems
- human faces
- training data
- face detection
- decision stumps
- decision trees
- multiclass classification
- majority voting
- feature representations
- classification trees
- support vector
- resource allocation
- attribute values
- linear classifiers
- ensemble methods
- training set
- face images
- test set
- classification models
- bayesian classifiers
- feature set
- weighted voting
- cost sensitive
- combining classifiers
- learning algorithm
- training samples
- facial expressions
- object detection
- base classifiers
- latent variables
- game theory
- classifier combination
- multi class
- base learners
- binary classification
- loss function
- multi agent
- prediction accuracy
- machine learning algorithms
- naive bayes