Backpropagation

When the output is wrong, how much is each weight to blame?

A tiny network, one training example, and every number shown. The table at the bottom is the heart of how all neural networks learn: for each weight, backprop multiplies a few local slopes together to get "how much does the error change if I nudge this weight?"

The network after 11557 updates

w1 = 0.384 w2 = 0.668 w3 = 0.483 w4 = 0.765 w5 = −3.933 w6 = −3.908 w7 = 2.902 w8 = 2.966 i1 0.050 i2 0.100 h1 0.607 h2 0.611 o1 0.015 o2 0.985 target 0.01 target 0.99 bias b1 = 0.35 output bias b2 = 0.6

Train it

0.5
0.000026total error E
0.0151, 0.9848outputs (want 0.01, 0.99)
0.00000.0752,8890.1495,7790.2248,6680.29811,557updates

1. Forward pass: compute the answer

net_h1 = w1·i1 + w2·i2 + b1 = 0.3838·0.05 + 0.6677·0.10 + 0.35 = 0.435958
h1 = σ(net_h1) = 0.607295    h2 = σ(0.450659) = 0.610796
net_o1 = w5·h1 + w6·h2 + b2 = −4.175589    o1 = 0.015134
net_o2 = w7·h1 + w8·h2 + b2 = 4.173841    o2 = 0.984840
E = ½(t1 − o1)² + ½(t2 − o2)² = 0.000013 + 0.000013 = 0.000026488

What to notice

Output weights are easy. w5's blame is just three numbers multiplied: how wrong o1 is, how steep o1's sigmoid is, and how big h1 was (a big input means the weight mattered more).

Hidden weights reuse that work. w1 can't see the error directly. Its blame arrives through both outputs: output 1's blame times w5, plus output 2's blame times w7. Then the same local slopes. That reuse is what makes backprop fast enough for billions of weights.

The check column agrees. "Nudge and measure" gives the same numbers without any calculus, but it needs two extra forward passes per weight. For an LLM that would be billions of passes per step; backprop needs one.

2. Backward pass: blame for every weight

WeightLocal slopes multiplied together (chain rule)∂E/∂wCheck (nudge)New value = w − 0.50 × ∂E/∂w
w5 (o₁ − t₁) 0.00513 × o₁(1 − o₁) 0.01490 × h₁ 0.60730 0.0000465 0.0000465 −3.93275 → −3.93277434
w6 (o₁ − t₁) 0.00513 × o₁(1 − o₁) 0.01490 × h₂ 0.61080 0.0000467 0.0000467 −3.90842 → −3.90844329
w7 (o₂ − t₂) −0.00516 × o₂(1 − o₂) 0.01493 × h₁ 0.60730 −0.0000468 −0.0000468 2.90182 → 2.90184480
w8 (o₂ − t₂) −0.00516 × o₂(1 − o₂) 0.01493 × h₂ 0.61080 −0.0000471 −0.0000471 2.96593 → 2.96595364
w1 δo₁·w5 + δo₂·w7 −0.00052 × h₁(1 − h₁) 0.23849 × i₁ 0.05000 −0.0000063 −0.0000063 0.38383 → 0.38383507
w2 δo₁·w5 + δo₂·w7 −0.00052 × h₁(1 − h₁) 0.23849 × i₂ 0.10000 −0.0000125 −0.0000125 0.66766 → 0.66767014
w3 δo₁·w6 + δo₂·w8 −0.00053 × h₂(1 − h₂) 0.23772 × i₁ 0.05000 −0.0000063 −0.0000063 0.48264 → 0.48263988
w4 δo₁·w6 + δo₂·w8 −0.00053 × h₂(1 − h₂) 0.23772 × i₂ 0.10000 −0.0000125 −0.0000125 0.76527 → 0.76527976

At the start with learning rate 0.5 you should see w5 → 0.35891648 and w1 → 0.14978072, the same as the original worked example. δo means (o − t)·o(1 − o), the output neuron's blame.

The code that just ran

Everything on this page came from calling this code, which lives at the top of this same file.

Show or hide the source
final class TinyNet
{
    public array $w = ['w1' => 0.15, 'w2' => 0.20, 'w3' => 0.25, 'w4' => 0.30,
                       'w5' => 0.40, 'w6' => 0.45, 'w7' => 0.50, 'w8' => 0.55,
                       'b1' => 0.35, 'b2' => 0.60];   // biases stay fixed, as in the original example
    public array $in = [0.05, 0.10];
    public array $target = [0.01, 0.99];

    public function forward(): array
    {
        $w = $this->w; [$i1, $i2] = $this->in; [$t1, $t2] = $this->target;
        $nh1 = $w['w1'] * $i1 + $w['w2'] * $i2 + $w['b1'];  $h1 = sigmoid($nh1);
        $nh2 = $w['w3'] * $i1 + $w['w4'] * $i2 + $w['b1'];  $h2 = sigmoid($nh2);
        $no1 = $w['w5'] * $h1 + $w['w6'] * $h2 + $w['b2'];  $o1 = sigmoid($no1);
        $no2 = $w['w7'] * $h1 + $w['w8'] * $h2 + $w['b2'];  $o2 = sigmoid($no2);
        $E1 = 0.5 * ($t1 - $o1) ** 2;
        $E2 = 0.5 * ($t2 - $o2) ** 2;
        return compact('i1', 'i2', 't1', 't2', 'nh1', 'h1', 'nh2', 'h2', 'no1', 'o1', 'no2', 'o2', 'E1', 'E2') + ['E' => $E1 + $E2];
    }

    public function loss(): float { return $this->forward()['E']; }

    // BACKPROPAGATION. Each gradient is a product of simple local slopes:
    //   (o − t)      how the error changes with the output
    //   o·(1 − o)    how the sigmoid output changes with its input
    //   h or i       how that input changes with the weight
    // Hidden weights get the SUM of the blame flowing back from both outputs.
    public function gradients(): array
    {
        $f = $this->forward(); $w = $this->w;
        $do1 = ($f['o1'] - $f['t1']) * $f['o1'] * (1 - $f['o1']);   // δ of output 1
        $do2 = ($f['o2'] - $f['t2']) * $f['o2'] * (1 - $f['o2']);   // δ of output 2
        $back1 = $do1 * $w['w5'] + $do2 * $w['w7'];                 // blame reaching h1
        $back2 = $do1 * $w['w6'] + $do2 * $w['w8'];                 // blame reaching h2
        $s1 = $f['h1'] * (1 - $f['h1']);
        $s2 = $f['h2'] * (1 - $f['h2']);

        $g = fn($factors) => ['factors' => $factors, 'value' => array_product(array_column($factors, 1))];
        return [
            'w5' => $g([['(o₁ − t₁)', $f['o1'] - $f['t1']], ['o₁(1 − o₁)', $f['o1'] * (1 - $f['o1'])], ['h₁', $f['h1']]]),
            'w6' => $g([['(o₁ − t₁)', $f['o1'] - $f['t1']], ['o₁(1 − o₁)', $f['o1'] * (1 - $f['o1'])], ['h₂', $f['h2']]]),
            'w7' => $g([['(o₂ − t₂)', $f['o2'] - $f['t2']], ['o₂(1 − o₂)', $f['o2'] * (1 - $f['o2'])], ['h₁', $f['h1']]]),
            'w8' => $g([['(o₂ − t₂)', $f['o2'] - $f['t2']], ['o₂(1 − o₂)', $f['o2'] * (1 - $f['o2'])], ['h₂', $f['h2']]]),
            'w1' => $g([['δo₁·w5 + δo₂·w7', $back1], ['h₁(1 − h₁)', $s1], ['i₁', $f['i1']]]),
            'w2' => $g([['δo₁·w5 + δo₂·w7', $back1], ['h₁(1 − h₁)', $s1], ['i₂', $f['i2']]]),
            'w3' => $g([['δo₁·w6 + δo₂·w8', $back2], ['h₂(1 − h₂)', $s2], ['i₁', $f['i1']]]),
            'w4' => $g([['δo₁·w6 + δo₂·w8', $back2], ['h₂(1 − h₂)', $s2], ['i₂', $f['i2']]]),
        ];
    }

    // Sanity check without any calculus: nudge the weight up and down a hair
    // and see how much the loss moves. Should match backprop almost exactly.
    public function numericalGradient(string $name, float $eps = 1e-5): float
    {
        $keep = $this->w[$name];
        $this->w[$name] = $keep + $eps; $up = $this->loss();
        $this->w[$name] = $keep - $eps; $down = $this->loss();
        $this->w[$name] = $keep;
        return ($up - $down) / (2 * $eps);
    }

    public function step(float $lr): void
    {
        foreach ($this->gradients() as $name => $g) $this->w[$name] -= $lr * $g['value'];
    }
}