The Devil is in the Details: Evaluating Limitations of Transformer-based Methods for Granular Tasks.
Brihi JoshiNeil ShahFrancesco BarbieriLeonardo NevesPublished in: CoRR (2020)
Keyphrases
- machine learning methods
- preprocessing
- empirical studies
- genetic algorithm
- significant improvement
- cross validation
- qualitative and quantitative
- artificial neural networks
- hidden markov models
- computational cost
- fuzzy logic
- data mining techniques
- high dimensional data
- classification method
- computationally expensive
- statistical methods