Fun with visual pattern recognition!

Aditya Patil (~aditya89)


4

Votes

Description:

The motto is to introduce concept of visual pattern recognition using Artificial neural networks. To show how a model learns from training examples and to use that model for recognising the handwritten patterns.

The talk consists of :

  1. Introduction to artificial neural network, Perceptrons and Sigmoid function.

  2. Architecture of neural networks .

  3. Simple network to classify handwritten digits.

  4. Implementation of gradient descent to learn the model.

  5. Implementing our network to classify digits.

  6. Limitations of Artificial neural network.

  7. Why "Convolutional Neural Networks" and "Capsule Networks" are preferred over Artificial neural network?

By the end of the session one will be able to understand what deep learning is and why it is necessary.

Prerequisites:

Basic python programming and maths.

Content URLs:

Content url will be provided after the session in the form of github repo.

Speaker Info:

I'm Aditya Patil , pursuing my career in Computer Science Engineering at Government Engineering College Aurangabad, Maharashtra. I have been working in data science from past year. Whilst conjured interest in Machine learning. I have knowledge of BIG DATA technologies such as Hadoop, Hive, Pig and SPARK.

Speaker Links:

https://www.linkedin.com/in/aditya-patil-a7b543152/

Section: Data science
Type: Talks
Target Audience: Beginner
Last Updated:

The comment is marked as spam.

EmmaSwan7618

Actually this problem is much complicated, the visually impaired need to be trained to create a mental map for independant navigation of such websites without having to read. One thing can be done, colour identification of the images can be

Aditya Patil (~aditya89)
The comment is marked as spam.

BelkinSetup
The comment is marked as spam.

edwardgayle

Login to add a new comment.