Recently, I do some literature research about CNN and find there is a concept of
end to end training
Such as the abstract in Fully Convolutional Networks for Semantic Segmentation
How to understand that? What is end-to-end training?
Solved – how to understand ‘end to end’ in deep learning
conv-neural-networkdeep learningmachine learning
Related Question
- Solved – a simplified version of fully convolutional network
- Deep Learning – What Does End to End Mean in Deep Learning Methods?
- Solved – Reference to learn how to interpret learning curves of deep convolutional neural networks
- Solved – Why is softmax output not a good uncertainty measure for Deep Learning models
- Solved – How to train the deep learning model on another similar yet different dataset
Best Answer
Traditionally, we extract pre-defined features before prediction.
However, hand-engineered features limit the potential performance as some of them are poor approximation of reality and some of them throw away some information.
End-to-end learning means that we replace the pipeline with a single learning algorithm so that it goes directly from the input to the desired output to overcome limitations of the traditional approach.
End-to-end learning system tend to do better when there is a lot of labeled data as the learning algorithm can somehow learn features by itself. When the training set is small, it tends to do worse than hand-engineered pipeline.