Machine Learning Unsupervised Backpropagation

0 votes

I'm having trouble with some of the concepts in machine learning through neural networks. One of them is backpropagation. In the weight updating equation,

delta_w = a*(t - y)*g'(h)*x

t is the "target output", which would be your class label, or something, in the case of supervised learning. But what would the "target output" be for unsupervised learning?

Can someone kindly provide an example of how you'd use BP in unsupervised learning, specifically for clustering of classification?

Thanks in advance.

Apr 5 in Machine Learning by Dev
• 6,000 points

1 answer to this question.

0 votes
The most frequent method is to train an autoencoder to produce outputs that are equal to the inputs. As a result, the network will strive to develop a representation that "compresses" the input distribution the best it can.
The derivatives of the error function are computed using backpropagation when training an artificial neural network with regard to the weights in the network. It's called thus because the "errors" are "propagating" backwards through the network. In this scenario, you'll need it because the ultimate error with regard to the target is determined by a function of functions (of functions ... depending on how many layers in your ANN.) The derivatives then allow you to tweak the variables to improve the error function, while the learning rate keeps things in check.
This isn't necessary in unsupervised methods. When using k-Means to minimize the mean squared error (MSE), for example, you may minimize the error directly at each step given the assignments; no gradients are required. The expectation-maximization (EM) approach is far more powerful and accurate than any gradient-descent based method in other clustering models, such as a combination of Gaussians.
answered Apr 7 by Nandini
• 5,480 points

Related Questions In Machine Learning

0 votes
1 answer

What is semi-supervised machine learning?

Hi@Ganesh, Semi-supervised machine learning is a combination of ...READ MORE

answered Jul 19, 2020 in Machine Learning by MD
• 95,340 points
0 votes
1 answer

Real world applications of Machine Learning

Few real-world applications of machine learning are  Have ...READ MORE

answered May 10, 2019 in Machine Learning by Jinu
0 votes
1 answer
0 votes
1 answer

What is clustering in Machine Learning?

Clustering is a type of unsupervised learning ...READ MORE

answered May 10, 2019 in Machine Learning by Shridhar
0 votes
1 answer

Machine Learning: Unsupervised Backpropagation

Backpropagation in unsupervised learning; probably the models ...READ MORE

answered Feb 23 in Machine Learning by Dev
• 6,000 points
0 votes
2 answers

How do neural networks used in AI and deep learning learn?

A neural network is a hardware or software ...READ MORE

answered Apr 8, 2019 in Others by SA
• 1,050 points
0 votes
1 answer

What is Unsupervised Learning?

Unsupervised Learning is the training of machine ...READ MORE

answered May 8, 2019 in Machine Learning by Alok
0 votes
1 answer

What is inductive bias in machine learning?

Inductive bias can be understood as an ...READ MORE

answered Feb 10 in Machine Learning by Nandini
• 5,480 points
Send OTP
webinar_success Thank you for registering Join Edureka Meetup community for 100+ Free Webinars each month JOIN MEETUP GROUP