Checking Whether More Same-Distribution Data Would Help
Question: You add a small set of examples from the target environment into training. The model performs well on those added examples, but it still performs poorly on a separate held-out subset from the same target environment. What conclusion does that support?
Sample answer: It suggests that training on more examples from that target environment would likely improve performance.
Key points:
- Supports the idea that additional data from the same target distribution would help.
- Shows the model can fit that distribution, but the current sample is too small to generalize well.
Rubric: The answer should say that more data from the target distribution is likely to improve generalization.
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
Testing a Data-Mismatch Hypothesis on a Shared Subset
A wildlife classifier has 8,000 camera-trap photos from reserve staff and 160,000 unrelated web photos. How should the reserve photos be used?
Combining In-Domain and Auxiliary Examples During Training
Adding target-distribution examples to training means the training set now includes data from the _____ distribution.
Match the training-set components with their distribution role in a model built for a specific target domain.
Order the steps for allocating product-review data between target and auxiliary sources.
Why can it help to include some target-distribution audio in both the training set and the training-dev set?
True or False: It can still be worthwhile to train on a dataset that comes from a different distribution than the dev/test set.
Strong results on source examples but weak results on deployment examples validate the hypothesis that more _____ data would help.
Match each training choice for a voice-command system to its likely effect.
Order the steps for training a defect classifier with mixed data sources.
Why Include a Small Amount of Target-Distribution Data in Training?
Using Target-Distribution Samples to Guide Training Decisions
Checking Whether More Same-Distribution Data Would Help