Auto encoders are a type of Artificial Neural Networks that are used to perform a task of data encoding (representation learning).

Auto encoders use the same input data for the input and output, Sounds crazy right!?

Auto encoders work by adding a bottleneck in the network. This bottleneck forces the network to create a compressed (encoded) version of the original input.

Auto encoders work well if correlations exists between input data (performs poorly if the all input data is independent)

Subscribe to my channel to get the latest updates, we will be releasing new videos on weekly basis:
[ Ссылка ]

Thanks and happy learning!

#autoencoders #deeplearning #machinelearning

свернуть