tute2_ans(2)

.pdf

School

Nanyang Technological University *

*We aren’t endorsed by this school

Course

MISC

Subject

Statistics

Date

Nov 24, 2024

Type

pdf

Pages

50

Uploaded by DukeLemur2884

Regression and Classification SC4001 – Tutorial 2
࠵? = ࠵? ! , ࠵? " , ࠵? # ࠵? 0.09, −0.44, −0.15 −2.57 0.69, −0.99, −0.76 −2.97 0.34, 0.65, −0.73 0.96 0.15, , 0.78, −0.58 1.04 −0.63, −0.78, −0.56 −3.21 0.96, 0.62, −0.66 1.05 0.63, −0.45, −0.14 −2.39 0.88, 0.64, −0.33 0.66
initial weights and biases: ࠵? = 0.77 0.02 0.63 , ࠵? = 0.0 Learning factor ࠵? = 0.01 ࠵? ! ࠵? " ࠵? # ࠵? +1 ࠵? = ࠵? ࠵? = ࠵?
Given a training dataset ࠵? $ , ࠵? $ $%! & Set learning parameter ࠵? Initialize ࠵? and ࠵? Repeat until convergence: For every training pattern ࠵? $ , ࠵? $ : ࠵? $ = ࠵? $ ࠵? + ࠵? ࠵? ← ࠵? + ࠵? ࠵? $ − ࠵? $ ࠵? $ ࠵? ← ࠵? + ࠵? ࠵? $ − ࠵? $ (a) SGD for the linear neuron :
SGD: learning factor ࠵? = 0.01 Shuffle the inputs. Epoch 1 ࠵? = 1 Apply ࠵? $ = 0.34 0.65 −0.73 , ࠵? $ = 0.96 ࠵? $ = ࠵? ࠵? & ࠵? + ࠵? = 0.34 0.65 −0.73 0.77 0.02 0.63 + 0.0 = −0.19 ࠵?. ࠵?. = ࠵? $ − ࠵? $ " = 0.96 + 0.19 " = 1.31 ࠵? ← ࠵? + ࠵? ࠵? $ − ࠵? $ ࠵? ࠵? = 0.77 0.02 0.63 + 0.01× 0.96 + 0.19 0.34 0.65 −0.73 = 0.78 0.03 0.62 ࠵? ← ࠵? + ࠵? ࠵? $ − ࠵? $ = 0.0 + 0.01× 0.96 + 0.19 = 0.01
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help