← All workshopsRead Learning From a Few Records at a Time

Optimisation and Training

Update a weight from a batch

BOTIE

Return the average of residual times input across the supplied batch.
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

Return the average of residual times input across the supplied batch.

Botie has aligned inputs and targets for a small training batch.

Inspect the component calculation
Mean(
  Numbers: Multiply(
    Value: Subtract(
      Starting value: Multiply(
        Value: Batch inputs
        Multiplier: Current weight
      )
      Value to remove: Batch targets
    )
    Multiplier: Batch inputs
  )
)
Examples and results
Batch inputs
  1. 01
  2. 12
Current weight
1
Batch targets
  1. 02
  2. 14
Required result
-2.5
Batch inputs
  1. 03
Current weight
2
Batch targets
  1. 09
Required result
-9
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

This model has one weight, no intercept, and half mean squared error. Numerical tests allow a difference of 0.001. Lists and structures must also have the right shape.

Your last tested pieces can start the full construction. An existing draft there keeps its own pieces.

Two training records ask for opposite changes

This check uses the pieces from “Let each batch record contribute a weight gradient”.

BOTIE

The two records produce gradients with opposite signs for the shared weight. Average their gradients for half squared error, apply one update, and return the new weight.
Batch inputs
  1. 01
  2. 12
Batch targets
  1. 04
  2. 11
Current weight
1
Learning rate
0.1

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 would replacing the mean gradient with a sum change the effect of batch size?

Match gradient reduction to loss reduction when the number of records changes.

This model has one weight, no intercept, and half mean squared error.

Work through the Python challenges →
Continue the courseRemembering the Direction of Earlier Updates →
← Residual Connections