Auto encoders are a type of unsupervised learning technique of artificial neural networks that are mainly used to perform a task of data encoding or we call Representation learning. Representation Learning known as Auto Encoders are the extension part of Deep Learning which uses a kind of reverse engineering with basic kind of wheel architectures. auto encoders basically work by applying a dimensionality reduction that's very similar to principal component analysis or PCH. and PCH simply tries to perform dimensionality reduction. This work focused on the Quality of Service (QoS) in terms of Accuracy of the Deep Representation Learning Model optimization analysis which is known as Auto Encoders of Deep Learning model. Auto Encoder model QoS is tested on TPU of Intel® Core™ i3-7100U CPU using TensorFlow 2.0. Results are encouraged.

Representation Learning, Auto Encoders, Deep Learning, TensorFlow, TPU, Artificial Neural Network, Unsupervised Training, QoS.