How much data is sufficient to learn high-performing algorithms?
Maria-Florina BalcanDan F. DeBlasioTravis DickCarl KingsfordTuomas SandholmEllen VitercikPublished in: CoRR (2019)
Keyphrases
- data sets
- data structure
- data mining techniques
- database
- small number
- data collection
- high quality
- raw data
- data sources
- noisy data
- original data
- data mining algorithms
- statistical analysis
- data processing
- data quality
- incomplete data
- worst case
- knowledge discovery
- training data
- search engine
- image data
- end users
- significant improvement
- missing values
- data analysis
- computational complexity
- spectral clustering
- complex data
- learning algorithm
- data reduction