Subtle Inverse Crimes: Naïvely training machine learning algorithms could lead to overly-optimistic results.
Efrat ShimronJonathan I. TamirKe WangMichael LustigPublished in: CoRR (2021)
Keyphrases
- machine learning algorithms
- benchmark data sets
- learning algorithm
- machine learning
- predictive accuracy
- learning problems
- decision trees
- machine learning methods
- learning tasks
- training and test data
- random forests
- machine learning models
- training examples
- law enforcement
- text categorization
- text classification
- input features
- machine learning systems
- transfer learning
- neural network
- machine learning approaches
- training set
- machine learning and data mining algorithms
- model selection