UDAMA: Unsupervised Domain Adaptation through Multi-discriminator Adversarial Training with Noisy Labels Improves Cardio-fitness Prediction.
Yu WuDimitris SpathisHong JiaIgnacio Perez-PozueloTomas I. GonzalesSøren BrageNicholas J. WarehamCecilia MascoloPublished in: CoRR (2023)
Keyphrases
- training process
- training data
- label noise
- training algorithm
- neural network
- prediction accuracy
- evolutionary algorithm
- prediction model
- training examples
- genetic algorithm
- training set
- active learning
- genetic programming
- prediction algorithm
- fitness function
- data sets
- labelled data
- multi layer perceptron
- low quality
- learning algorithm
- supervised learning
- pairwise