Procedure for Training a Child to Identify a Cat using 10,000 Example Cats

1
Procedure for Training a Child to Identify a Cat
using 10,000 Example Cats

• For Cat_index ? 1 to 10000
• 1. Show cat and describe catlike features
  (Cat_index)
• 2. Child adjusts biological neural network in
  response to receiving the features of example
  cat Cat_index
• 3. Cat_index ? Cat_index 1

Procedure for Testing a Trained Childs ability
to Identify a new Cat
1. Show new cat and describe catlike features
2. Child processes features with biological
neural network in response to receiving the
features of new example cat 3. Output of
biological neural network indicates weather or
not new example is a cat
2
Smoothing function for converting the output of a
neuron into the range 0,1
3
Forward Pass Computations through a
Back-Propagation Neural Network with three layers
having 4, 6, and 8 nodes

• INPUT input(1),input(2),input(3),input(4)
• For i ? 1 to 6
• middle_in (i) ? 0
• For j ? 1 to 4
• middle_in (i) lt middle_in (i) weight(j,i)
  inp8ut (j)
• middle_out (i) ? Fermi (middle_in(i))
• For k ? 1 to 8
• output (k) ?0
• For i ? 1 to 6
• output (k) ? output (k) weight (i,k)
  middle_out (i)
• INPUT known_true_value (k)
• error (k) ? known_true_value (k) output (k)

4
General Procedure for training a neural network,
then testing it on new examples

• INPUT known true values for each example
• For i ? 1 to number_of_examples_in_input_set
• INPUT numbers that measure values of input
  features for this example
• INPUT known true classification values for this
  example
• Do forward neural net computation to get
  outputs
• Compute error by subtracting known true values
  from outputs
• Set error_tolerance_threshold
• Repeat until error tolerance lt
  error_tolerance_threshold
• Do backpropagation for an epoch and adjust
  weights