Finding Order in Chaos: A Novel Data Augmentation Method for Time Series in Contrastive Learning.
Berken Utku DemirelChristian HolzPublished in: NeurIPS (2023)
Keyphrases
- synthetic data
- prior knowledge
- missing values
- statistical methods
- missing data
- test data
- background knowledge
- preprocessing
- input data
- multivariate time series
- image data
- supervised learning
- learning process
- noisy data
- objective function
- spatial data
- unsupervised learning
- data sets
- temporal data
- bayesian methods
- information loss
- data mining tasks
- data collection
- learning environment
- training samples
- clustering method
- learning tasks
- data structure
- reinforcement learning
- similarity measure
- learning algorithm
- method finds
- rare events
- stock data
- non stationary
- labeled data
- probability distribution
- data sources
- active learning
- k means
- pairwise
- feature selection
- machine learning
- data mining