Activation

Activation and weights are the two most important concepts for abstract neural network models, where activations are dynamic state variables representing neural activity that are communicated to other neurons over weighted synaptic connections. In a standard rate-code activation model, a single real-valued (floating point) number is used for the activation value. Biological neurons instead communicate via discrete action potentials or spikes.

Activation pages

• • Adaptation
• • Axon
• • Inhibition
• • Neuron bayesian
• • Neuron channels
• • Neuron dendrites
• • Neuron detector
• • Neuron electrophysiology
• • Neuron equilibrium potential
• • Rate code activation