40 Questions
Why are different weights assigned to the inputs in an artificial neuron?
To emphasize the relative importance of the different inputs
What is the mathematical form of the operation performed by an artificial neuron?
$f(x_1w_1 + x_2w_2 + x_3w_3 + b)$
What is the key difference between an artificial neuron and linear regression?
Both b and c
How are artificial neurons arranged in an Artificial Neural Network (ANN)?
In multiple layers, with connections only between adjacent layers
What is the role of the bias term in an artificial neuron?
It is used to shift the activation function to the left or right
What is the purpose of the activation function in an artificial neuron?
All of the above
What determines the number of neurons in the output layer in a neural network?
The number and type of problem to be solved
In binary classification, how many neurons are typically present in the output layer?
One
What is the purpose of introducing non-linearity in neural networks?
To introduce a nonlinear transformation to learn complex patterns
Which function scales the input value between 0 and 1 in neural networks?
Sigmoid function
What is the range of values output by the Tanh function in neural networks?
-1 to 1
In a regression problem, how many neurons are typically found in the output layer of a neural network?
One
What is the function of a synapse in a neuron?
It is where information is transmitted between neurons.
How are inputs to a neuron weighted before being summed in the cell body?
They are strengthened or weakened based on their importance.
What is the role of the soma in a neuron?
Processing summed inputs and sending them through axons.
In artificial neurons, what happens to the inputs received before they are processed?
They are weighted based on their importance.
Which of the following best describes the activation function of a neuron?
It introduces non-linearity into the network.
What distinguishes neurons in artificial neural networks from linear regression models?
Neurons in artificial neural networks can handle non-linear relationships, unlike linear regression models.
Artificial neurons are arranged by layers.
True
Nuerons in the same layer do not have any connections.
True
which one is not a typical ANN layers?
Forward Layer
The number of neurons in the input layer is the number of inputs we feed to the network
True
There is no computaion in the input layer. it is just used for passing information from the outside world to the network.
True
Any layer between the inpout layer and the output layer is called ___________.
Hidden Layer
The input layer identifies the pattern in the dataset.
False
__________ identifies the pattern in dataset and is responsible for deriving complex relationships between input and output.
Hidden Layer
The network is called a ___________________ when we have many hidden layers.
Deep Nueral Network (DNN)
The number of neurons in the output layer is based on the number and the types of the problems to be solved.
True
_______________ is used to introduce non-linearity in neural networks.
Activation Function
The aim of the activation function is to introduce a nonlinear transformation to learn the complex underlying patterns in data.
True
The _____________ scales the input values between 0 and 1.
Sigmoid Function
Sigmoid Function = 1 / (1+ exp(-x))
True
The ____________ outputs the values between -1 and +1
Hyperbolic tangent (tanh) Function
____________ outputs the values from 0 yo infinity.
ReLU (Rectified Linear Unit) Function
the sang for being zero for all negatives values is a problem called ________ and a neuron is said to be dead if it always outputs zero.
dying ReLU
_____________ is a variant of the ReLU function that solves the dying ReLU problem. The value of alpha typically se to 0.01
Leaky ,ReLU Function
ELU (Exponential Linear Unit) is like Leaky ReLU has a small slope for negative values.
True
__________ is basically the generalization of the sigmoid functions.
The Softmax Function
The Softmax Functions is usually applied to the final layer of the network and while performing multi-class classification tasks. It gives the probabilities of each class for being output and thus , the sum of softmax values will always equal 1.
True
softmax function = exp(x) / sum(exp(x)) [0,j]
True
Learn how artificial neurons work by understanding how inputs are multiplied by weights and summed together. Discover the importance of assigning different values to weights based on the significance of the inputs in calculating the output. Explore how weights are used to strengthen inputs in artificial neural networks.
Make Your Own Quizzes and Flashcards
Convert your notes into interactive study material.
Get started for free