vision_tasks.rst

Self-supervised Learning

This section implements popular contrastive learning tasks used in self-supervised learning.

Note

We rely on the community to keep these updated and working. If something doesn't work, we'd really appreciate a contribution to fix!

---

FeatureMapContrastiveTask

This task compares sets of feature maps.

In general the feature map comparison pretext task uses triplets of features. Here are the abstract steps of comparison.

Generate multiple views of the same image

x1_view_1 = data_augmentation(x1)
x1_view_2 = data_augmentation(x1)

Use a different example to generate additional views (usually within the same batch or a pool of candidates)

x2_view_1 = data_augmentation(x2)
x2_view_2 = data_augmentation(x2)

Pick 3 views to compare, these are the anchor, positive and negative features

anchor = x1_view_1
positive = x1_view_2
negative = x2_view_1

Generate feature maps for each view

(a0, a1, a2) = encoder(anchor)
(p0, p1, p2) = encoder(positive)

Make a comparison for a set of feature maps

phi = some_score_function()

# the '01' comparison
score = phi(a0, p1)

# and can be bidirectional
score = phi(p0, a1)

In practice the contrastive task creates a BxB matrix where B is the batch size. The diagonals for set 1 of feature maps are the anchors, the diagonals of set 2 of the feature maps are the positives, the non-diagonals of set 1 are the negatives.

pl_bolts.losses.self_supervised_learning.FeatureMapContrastiveTask

---

Context prediction tasks

The following tasks aim to predict a target using a context representation.

CPCContrastiveTask

This is the predictive task from CPC (v2).

task = CPCTask(num_input_channels=32)

# (batch, channels, rows, cols)
# this should be thought of as 49 feature vectors, each with 32 dims
Z = torch.random.rand(3, 32, 7, 7)

loss = task(Z)

pl_bolts.losses.self_supervised_learning.CPCTask