You trained a model that performs exceptionally well on the training data but poorly on the test data. What could be the issue, and how would you address it?
- Increase complexity
- Increase dataset size
- Overfitting, add regularization
- Reduce complexity
The issue is likely overfitting, where the model has learned the training data too well, including its noise and anomalies. Adding regularization would help to constrain the model and make it generalize better to unseen data.
Loading...
Related Quiz
- When dealing with high-dimensional data, which of the two algorithms (k-NN or Naive Bayes) is likely to be more efficient in terms of computational time?
- To avoid overfitting in large neural networks, one might employ a technique known as ________, which involves dropping out random neurons during training.
- In reinforcement learning, the agent learns a policy which maps states to ________.
- In Machine Learning, models learn from data and make predictions, whereas in Deep Learning, models can automatically learn representations from data through _________.
- In a high-dimensional dataset, how would you decide which kernel to use for SVM?