Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -36,9 +36,13 @@ We provide four splits:
|
|
36 |
- `test_mixed`: 20'000 images, consisting of the (shuffled) concatenation of our ambiguous `test` test and the nominal mnist test set by LeCun et. al.,
|
37 |
- `train_mixed`: 70'000 images, consisting
|
38 |
|
|
|
|
|
|
|
|
|
39 |
For research targeting explicitly aleatoric uncertainty, we recommend training the model using `train_mixed`.
|
40 |
Otherwise, our `test` set will lead to both epistemic and aleatoric uncertainty.
|
41 |
-
|
42 |
|
43 |
### Assessment and Validity
|
44 |
For a brief discussion of the strength and weaknesses of this dataset,
|
|
|
36 |
- `test_mixed`: 20'000 images, consisting of the (shuffled) concatenation of our ambiguous `test` test and the nominal mnist test set by LeCun et. al.,
|
37 |
- `train_mixed`: 70'000 images, consisting
|
38 |
|
39 |
+
Note that the ambiguous train images are highly ambiguous (i.e., the two classes have very similar ground truth likelihoods),
|
40 |
+
the training set images allow for more unbalanced ambiguity.
|
41 |
+
This is to make the training set more closely connected to the nominal data, while still keeping the test set clearly ambiguous.
|
42 |
+
|
43 |
For research targeting explicitly aleatoric uncertainty, we recommend training the model using `train_mixed`.
|
44 |
Otherwise, our `test` set will lead to both epistemic and aleatoric uncertainty.
|
45 |
+
In related literature, such 'mixed' splits are sometimes denoted as *dirty* splits.
|
46 |
|
47 |
### Assessment and Validity
|
48 |
For a brief discussion of the strength and weaknesses of this dataset,
|