=MMULT(HiddenActivation, W2) + b2
=AVERAGE((Predictions - TargetData)^2)
| | A | B | C | D | E | F | G | H | I | J | K | L | M | |-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----| | 1 | | A | B | Y | | W1 | | | b1 | | W2 | | b2 | | 2 | | | | | | col1| col2| | | | | | | | 3 | | 0 | 0 | 0 | | 0.5 | -0.6| | 0.1 | | 0.4 | | 0.2 | | 4 | | 0 | 1 | 1 | | 0.7 | 0.2 | | -0.2| | -0.3| | | | 5 | | 1 | 0 | 1 | | | | | | | | | | | 6 | | 1 | 1 | 0 | | | | | | | | | |
While this example demonstrates a simple neural network in Excel, it's essential to note that:
Here is what I learned about the soul of machine learning.
To follow this, you will need a Microsoft 365 subscription with enabled. Excel version: 2026 Excel (or later).
| Problem | Likely Cause | Solution | |---------|--------------|----------| | #VALUE! in matrix multiplication | Mismatched array dimensions | Verify that the number of columns in the first matrix equals the number of rows in the second. Use =TRANSPOSE() if needed. | | Sigmoid output always near 0 or 1 | Weights or inputs are too large | Smaller initial weights (random between –0.5 and 0.5) usually help. | | Training doesn’t reduce loss | Learning rate too high or too low | Start with learning_rate = 0.1 and adjust. Also check that backpropagation sign is correct. | | Excel becomes very slow | Too many array formulas or iterative calculations | Switch calculation mode to manual while iterating, then press F9 to recalculate only when needed. |
: You can type a prompt like: "Create a neural network to predict [Target] based on [Features] in Sheet1 using Python." .
Use the new Python-in-Excel capability to calculate gradients and update weights. Example PY() snippet: model.fit(X, y)
=MMULT(HiddenActivation, W2) + b2
=AVERAGE((Predictions - TargetData)^2)
| | A | B | C | D | E | F | G | H | I | J | K | L | M | |-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----| | 1 | | A | B | Y | | W1 | | | b1 | | W2 | | b2 | | 2 | | | | | | col1| col2| | | | | | | | 3 | | 0 | 0 | 0 | | 0.5 | -0.6| | 0.1 | | 0.4 | | 0.2 | | 4 | | 0 | 1 | 1 | | 0.7 | 0.2 | | -0.2| | -0.3| | | | 5 | | 1 | 0 | 1 | | | | | | | | | | | 6 | | 1 | 1 | 0 | | | | | | | | | | build neural network with ms excel new
While this example demonstrates a simple neural network in Excel, it's essential to note that:
Here is what I learned about the soul of machine learning. | Problem | Likely Cause | Solution |
To follow this, you will need a Microsoft 365 subscription with enabled. Excel version: 2026 Excel (or later).
| Problem | Likely Cause | Solution | |---------|--------------|----------| | #VALUE! in matrix multiplication | Mismatched array dimensions | Verify that the number of columns in the first matrix equals the number of rows in the second. Use =TRANSPOSE() if needed. | | Sigmoid output always near 0 or 1 | Weights or inputs are too large | Smaller initial weights (random between –0.5 and 0.5) usually help. | | Training doesn’t reduce loss | Learning rate too high or too low | Start with learning_rate = 0.1 and adjust. Also check that backpropagation sign is correct. | | Excel becomes very slow | Too many array formulas or iterative calculations | Switch calculation mode to manual while iterating, then press F9 to recalculate only when needed. | | | Sigmoid output always near 0 or
: You can type a prompt like: "Create a neural network to predict [Target] based on [Features] in Sheet1 using Python." .
Use the new Python-in-Excel capability to calculate gradients and update weights. Example PY() snippet: model.fit(X, y)
