← Lessons

quiz vs the machine

Gold1390

Machine Learning

Weight Initialization Strategies

Why the starting weights decide whether training even begins.

4 min read · core · beat Gold to climb

Weight Initialization Strategies

Before the first gradient step, weights need values. Set them poorly and signals either die or explode immediately, so initialization is a quiet but decisive choice.

Bad starting points

  • All zeros makes every neuron identical, so they learn the same thing forever.
  • Weights too large saturate activations and explode gradients.
  • Weights too small shrink the signal until learning stalls.

Principled schemes

The fix is to scale random weights by the layer width so variance stays stable across layers. Xavier or Glorot initialization targets tanh style activations by accounting for both input and output sizes. He initialization scales for ReLU networks, which discard half their inputs and so need slightly larger weights to preserve variance.

Why it matters

Good initialization places the network in a regime where activations and gradients have sensible magnitudes from step one. That lets the optimizer make progress instead of spending epochs recovering from a doomed start. Combined with normalization, modern schemes make very deep training stable from the first batch.

Key idea

Initialization scales random weights to the layer width so variance stays stable and training can start in a healthy regime.

Check yourself

Answer to earn rating on the learn ladder.

1. Why is initializing all weights to zero a problem?

2. He initialization is designed primarily for which activation?