single perceptron

TRUE OF FALSE

1) A single perceptron can compute the XOR function. [ ]

2) A single Threshold-Logic Unit can realize the AND function. [     ]

3) A perceptron is guaranteed to perfectly learn a given linearly separable function within a finite number of training steps.  [  ]

4) The more hidden-layer units a BPN has, the better it can predict desired outputs for new inputs that it was not trained with. [   ]

5) A three-layer BPN with 5 neurons in each layer has a total of 50 connections and 50 weights. [  ]

6) The backpropagation learning algorithm is based on the gradient descent method.[  ]

7) An epoch is when all of the data in the training set is presented to the neural network once. [     ]

8) Training set is a set of pairs of output patterns with corresponding input patterns. [  ]

9) ANN’s “designed to detect, and respond to, the presence of features in an input pattern vector that is presented as a dynamic pattern to the network.” [     ]

10) In backpropagation learning, we should start with a small learning parameter y and slowly increase it during the learning process. [     ]

11) The line equation for the following (w1 = -0.4, w2 = -0.5,  q= -0.3) is x2 = -0.4/-0.5 x + -0.3/-0.5 . [     ]

12) Delta Rule is capable of training all the weights in multilayer nets with no a priori knowledge of the training set. [     ]

13) Backpropagation algorithm is a generalized delta rule. [     ]

14) The sigmoid function uses the net amount of excitation as its argument. [     ]

15) The backpropagation algorithm is used to find the global minimum of the error function. [     ]

16) Normalization fumction is (Xi – min(X) / (Max (X) –Min(X)) [     ]

17) Neurons in the hidden layer can’t be observed through the input/output behaviour of the network.   [   ]

18) Multi-layer feed-forward networks can learn any function provided they have enough units and time to learn [    ]