CS 451 Quiz 8
Neural Nets 1
* Required
Email address
*
Your email
One can build a reliable car detector for 100x100 grayscale images by using logistic regression, where the input features are simply the 10,000 pixel values of the input image
*
1 point
True
False
Neural Networks are among the best machine learning techniques today. When were they first popular?
*
1 point
1950s
1980s
early 2000s
The fact that the animal (and human) brains can be rewired so that, say, the auditory cortex learns to see, is evidence for
*
1 point
audiovisual confusability
the phylogenetic tree hypothesis
the "one learning algorithm" hypothesis
A "neuron" (aka node) in a neural network is
*
1 point
an accurate model of a human neuron
inspired by human neurons but much simpler
inspired by the passing of genetic traits from parent to offspring
A single node in a neural network computes the exact same function as logistic regression
*
1 point
True
False
Consider a multilayer neural net, in which two successive layers each have 10 nodes. The parameters (weights) controlling the function mapping between these two layers are stored in a
*
1 point
10 x 1 vector
10 x 2 matrix
10 x 10 matrix
10 x 11 matrix
In a vectorized implementation of neural nets, the computation done by each layer involves a matrix multiplication, followed applying the logistic function g() to a vector
*
1 point
True
False
The Boolean AND function can be computed by a single neuron.
*
1 point
True
False
What function does the following neuron compute?
*
1 point
AND
OR
XOR
Multilayer neural nets are an attractive alternative to logistic regression with lots of polynomial features.
*
1 point
True
False
Submit
This content is neither created nor endorsed by Google.
Report Abuse

Terms of Service

Additional Terms
Forms