8
$\begingroup$

Shortly about deep learning (for reference):

Deep learning is a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers, composed of multiple linear and non-linear transformations.

Various deep learning architectures such as deep neural networks, convolutional deep neural networks, deep belief networks and recurrent neural networks have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks.


My question:

Can deep neural networks or convolutional deep neural networks be viewed as ensemble-based method of machine learning? Or it is different approaches?

$\endgroup$
3
$\begingroup$

You should think of them as different approaches. A deep neural net is a single independent model, whereas ensemble models are ensembles of many independent models.

The primary connection between the two is dropout, a particular method of training deep neural nets that's inspired by ensemble methods.

$\endgroup$
2
$\begingroup$

Deep neural networks could - in principle - be a component of an ensemble of machine learning algorithms, yes. Ensemble method basically just means use multiple algorithms and combining their output somehow.

Other than that, I don't see any special connection between deep learning and the idea of ensemble methods. DL is just one more tool in the toolkit.

$\endgroup$
0
$\begingroup$

I'm not sure the answers here properly address the question.

Ensemble methods are layers of methods applied to a data pipeline. Ensemble methods can contain the same sort of model in succession, or different models.

If you think of a multilayer neural network, essentially you are passing a data pipe through a succession of models. Each layer has its own "function", which can be the same function in succession or different ones. Each layer is essentially a regression, trying to figure out the proper coefficients. Tie this with modifier layers, such as batch normalization, regularization, etc, and you essentially have a pipeline of different models that you are pushing data through.

So in this view, a deep NN can be viewed as an ensemble of layers performing a data transformation.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.