Learn Before
Should Extra Labeled Data Be Added to a High-Variance Bird Call Classifier?
Case context: You are building an audio classifier that identifies bird species from short sound clips. The model has high variance, and you have found a large archive of unlabeled field recordings that could be labeled and added to training. At the same time, the compute budget for training and retraining is almost gone.
Question: Using the standard guidance for reducing variance, decide whether adding this new data is a practical next step right now, and explain the reason.
Sample answer: It is not a practical next step right now. Even though there is a large supply of additional data that could be labeled, the team does not have enough compute left to train effectively on it. This remedy only works when both conditions are met: sufficient extra data and enough computational capacity to use it.
Key points:
- Recognize that the model has high variance.
- Note that more data is available.
- Explain that the plan is not practical because compute is insufficient.
- State that both extra data and adequate compute are required for this approach.
Rubric: The student must say that adding training data is not currently a practical solution. The explanation must make clear that the available data condition is satisfied, but the compute requirement is not satisfied because the training budget is exhausted.
0
1
Tags
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
What effect does adding more training data usually have on a model that has high variance?
True or False: When a model shows high variance, collecting additional training examples is a dependable way to reduce overfitting.
Adding more training data can usually reduce variance without changing bias much
A classifier has high variance. What is usually the simplest and most dependable first fix?
True or False: Adding more training examples usually lowers variance and also lowers bias.
Fill in the blank: One of the simplest and most _____ ways to address high variance is to add more training examples.
Match each concept to its role when extra training data is used to reduce variance.
Order the steps a practitioner follows before using more training examples to address overfitting.
When is collecting additional training data a practical way to reduce variance?
True or False: For a model with high variance, collecting more training examples should be considered only after every other remedy has already been tried.
Adding more training examples usually reduces _____ without changing bias much.
Match each description to Andrew Ng’s view of adding more training data as a remedy for high variance.
Order the checks used to confirm that more training data reduced variance without raising bias
When Additional Data Helps Reduce Variance
Should Extra Labeled Data Be Added to a High-Variance Bird Call Classifier?
How More Training Data Affects Bias