Learn Before
Concept

End-to-End Learning Needs Abundant Labeled Input-Output Data

In the autonomous-driving example, a pure end-to-end approach would need a large dataset of image and steering-direction pairs. Collecting such data is time-consuming and expensive because it would require specially instrumented cars and a huge amount of driving to cover a wide range of scenarios, making the end-to-end system difficult to train.

0

1

Updated 2026-05-26

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy