generalization error
[/ˌdʒɛnərəlaɪˈzeɪʃən ˈɛrər/]
nounpl: generalization errors
erro de generalização
1. In machine learning, the difference between a model's performance on training data and its performance on unseen test data; measures how well a trained model generalizes to new, previously unseen examples
The model achieved 95% accuracy on training data but only 78% on test data, indicating a significant generalization error.
O modelo alcançou 95% de precisão nos dados de treinamento, mas apenas 78% nos dados de teste, indicando um erro de generalização significativo.
2. The failure of a machine learning model to perform equally well on new data due to overfitting or underfitting
High generalization error suggests the model may be overfitted to the training set.
Um alto erro de generalização sugere que o modelo pode estar sobreajustado ao conjunto de treinamento.
This is a technical term primarily used in machine learning, data science, and artificial intelligence fields. It is equally understood in both Brazilian Portuguese and European Portuguese academic and professional contexts. The term is relatively recent in Portuguese, with many professionals using the English term directly or the Portuguese translation interchangeably in technical discussions.
Look up more words on Fala2Me
The free English-Portuguese dictionary with real Brazilian accents, NYC slang, conjugator and more
Open Fala2Me →