MLP and its variants can be used to build a reconstruction model such as conventional autoencoders including many variants and GAN, for example.

I wonder if SOM can be a candidate to replace the role of multi-layer perceptrons in this manner.

The wiki page says 'The goal of learning in the self-organizing map is to cause different parts of the network to respond similarly to certain input patterns.' Thus I thought the above attempt would be conceptually possible. But how in the details?


Here is a response, hopefully not too simplistic. A single-layer SOM can perform some tasks that require multiple layers of perceptrons. For example, the XOR problem can be solved by a SOM, but not by a single layer perceptron.

In terms of reconstructive models, SOM has a codebook vector like a vector quantization method. A SOM essentially encodes each input as the best matching value in the codebook. As the SOM units train, they learn to fit the distribution, such that the codebook vectors approach the training data, while maintaining a topological arrangement (nodes close to each other are similar to one another).

So I think my answer is that the SOM already achieves something similar to an autoencoder, if you use the codebook values as the output. It might even work better for certain classes of data (perhaps strongly topological?), but my feeling is that, for many general tasks, it may be difficult to tune the SOM to perform as well as the others. On the other hand, SOM could be a good way to understand or explore the topology of your data.

| cite | improve this answer | |

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.