overfitting
[/ˌoʊvərˈfɪtɪŋ/]
noun
sobreajuste
1. A machine learning phenomenon where a model learns the training data too well, including its noise and peculiarities, resulting in poor performance on new, unseen data
The neural network suffered from overfitting because it memorized the training examples instead of learning generalizable patterns.
A rede neural sofreu com sobreajuste porque memorizou os exemplos de treinamento em vez de aprender padrões generalizáveis.
2. The condition where a statistical model captures random fluctuations in the data rather than the underlying relationship
To prevent overfitting, researchers use techniques like cross-validation and regularization.
Para evitar sobreajuste, pesquisadores utilizam técnicas como validação cruzada e regularização.
Overfitting is a fundamental concept in machine learning and data science communities worldwide. In Brazil and Portugal, both 'sobreajuste' and the English term 'overfitting' are commonly used in academic and professional settings. The term reflects the globalized nature of AI and ML terminology, with technical professionals often code-switching between English and Portuguese when discussing complex algorithmic concepts.
Look up more words on Fala2Me
The free English-Portuguese dictionary with real Brazilian accents, NYC slang, conjugator and more
Open Fala2Me →