Receptive Field

/rɪˈsɛptɪv fiːld/

Receptive Field: The region of input space that affects a particular neuron's activation in a neural network. In CNNs, receptive field grows with each layer. Deeper neurons see larger input regions. Determined by filter size, stride, and network depth. Critical for understanding what information neurons can access.

In italiano: Receptive FieldLegge III · Come una rete impara dai dati

In CNNs, receptive field grows with each layer. Deeper neurons see larger input regions. Determined by filter size, stride, and network depth. Critical for understanding what information neurons can access.

Examples

  • 3x3 filter has 3x3 receptive field
  • Deeper layers see larger regions
  • Dilated convolutions increase receptive field