Scribd
Upload
58 views2 pages

Perceptron

A perceptron is a basic machine learning algorithm that takes a set of input features with associated weights, computes a sum of the weighted inputs, passes the sum through an activation function to produce a binary output, and adjusts its weights during training to minimize error between its predictions and actual targets. It is a simple neural network unit that can only solve linearly separable problems but paved the way for more advanced neural architectures.

Uploaded by

Marcelo Zanzotti
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
58 views2 pages

Perceptron

A perceptron is a basic machine learning algorithm that takes a set of input features with associated weights, computes a sum of the weighted inputs, passes the sum through an activation function to produce a binary output, and adjusts its weights during training to minimize error between its predictions and actual targets. It is a simple neural network unit that can only solve linearly separable problems but paved the way for more advanced neural architectures.

Uploaded by

Marcelo Zanzotti
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

A perceptron is a basic building block of artificial neural networks and is one of the simplest

machine learning algorithms used for supervised learning of binary classifiers. It was developed
by Frank Rosenblatt in the late 1950s.

Here are the key components and concepts associated with a perceptron:

Input Features: A perceptron takes a set of input features, typically represented as a vector,
where each feature is assigned a weight.

Weights: Each input feature is associated with a weight that represents its importance or
contribution to the model's decision. These weights are learned during the training process.

Summation: The inputs are linearly combined with their corresponding weights, and the sum is
computed as follows:

where $n$ is the number of input features.

Activation Function: The sum is then passed through an activation function (also known as a
step function or threshold function). The activation function helps determine the output of the
perceptron. The most commonly used activation function is the Heaviside step function:

1, & \text{if Sum} \geq \text{Threshold} \\

0, & \text{if Sum} < \text{Threshold}

\end{cases}$$

The threshold is often considered as an additional weight (a bias) that can be learned along
with the other weights.

Learning: During the training phase, the perceptron's weights are adjusted to minimize the
error between its predicted output and the actual target output for a set of training examples.
This process typically uses a supervised learning algorithm like the perceptron learning rule or
gradient descent.

Output: The perceptron produces a binary output (0 or 1) based on the result of the activation
function. It can be seen as making a binary decision, such as classifying input data into one of
two classes.
A single perceptron can only model linearly separable functions, which means it can solve
problems where a straight line (or hyperplane in higher dimensions) can separate the classes in
the feature space. However, for more complex problems that are not linearly separable,
multiple perceptrons are typically combined into layers to form a multi-layer perceptron (MLP),
also known as a feedforward neural network. This allows neural networks to learn complex
non-linear relationships in data.

Perceptrons were a foundational concept in the history of neural networks and paved the way
for more advanced neural network architectures like feedforward neural networks,
convolutional neural networks (CNNs), and recurrent neural networks (RNNs). These more
complex neural networks are capable of solving a wide range of machine learning tasks,
including image recognition, natural language processing, and more.

You might also like