← All workshopsRead Scoring Connections We Have Not Observed

Graphs and Relational Learning

Score a possible connection

BOTIE

Take their dot product and pass the logit through sigmoid.
Ready to test
0 passed · 0 failed · 2 not run

An empty floor. Start with the information your machine needs.

Received result

Your answer will arrive here.

Examples, experiments and workspace controls
Live values

Connect a tool's inputs to see what it produces. You can try a branch before connecting it to Result.

Try your own inputs

Use a copy of the example to try your own data. Keep the problem's assumptions in mind, and predict the effect of a change before checking the output. You can also add another branch to compare two ways of doing the calculation.

Start at an output arrow, then click or drag to an input. Drag empty space to pan. Delete removes a selection. Ctrl/Cmd+Z undoes a change.
Context for this challenge

Take their dot product and pass the logit through sigmoid.

Botie has representations for two nodes and wants a bounded score for their possible connection.

Take the embedding dot product as a logit and pass it through sigmoid.

Examples and results
First node embedding
  1. 01
  2. 10
Second node embedding
  1. 02
  2. 11
Required result
0.8808
First node embedding
  1. 01
  2. 10
Second node embedding
  1. 0-2
  2. 11
Required result
0.1192
Try a fresh construction

This clears this draft, including its undo history. Earlier successful attempts stay recorded. Build again without opening hints to record an independent attempt.

Stuck? Inspect a worked construction

The tools below form one possible solution. You can assemble it, follow its signals, and undo that change to return to your own construction.

Assumptions and test precision

Supplied embeddings must be learned without held-out target edges. The score is not automatically calibrated. Numerical tests allow a difference of 0.001. Lists and structures must also have the right shape.

Several coordinates contribute to a possible link

This check uses the pieces from “Decode one candidate node pair”.

BOTIE

Botie has two learned three-coordinate node vectors. Calculate the link score by applying sigmoid to their dot product. Keep the negative coordinate contribution when forming the score.
First node embedding
  1. 01
  2. 12
  3. 2-1
Second node embedding
  1. 02
  2. 10.5
  3. 22

Use a number, text, or an array matching the requested result. For a matrix, put each row in brackets. Saving your prediction runs your current construction on these inputs and checks the earlier examples too.

Build a construction in the workspace above first. This case will test those pieces.

Why must held-out target edges be excluded when constructing or learning the node embeddings?

A link score can leak its answer through the graph used to make its representations.

Supplied embeddings must be learned without held-out target edges. The score is not automatically calibrated.

Work through the Python challenges →
Continue the courseSeparating a Useful Prediction From a Causal Claim →
← Learning From a Node and Its Neighbours Together