# To Ensemble or Not Ensemble: When does End-To-End Training Fail?

## TL;DR

This paper investigates the conditions under which end-to-end training of network ensembles fails, revealing failure cases, optimal intermediate strategies, and connections to Dropout and ensemble diversity.

## Contribution

It provides a detailed analysis of E2E training failures for ensembles, introducing a blending strategy and highlighting when partial or no E2E training is optimal.

## Key findings

- E2E training can fail for over-parameterized ensembles.
- Optimal training may lie between independent and fully E2E approaches.
- Links between E2E training, Dropout, and ensemble diversity are uncovered.

## Abstract

End-to-End training (E2E) is becoming more and more popular to train complex Deep Network architectures. An interesting question is whether this trend will continue-are there any clear failure cases for E2E training? We study this question in depth, for the specific case of E2E training an ensemble of networks. Our strategy is to blend the gradient smoothly in between two extremes: from independent training of the networks, up to to full E2E training. We find clear failure cases, where over-parameterized models cannot be trained E2E. A surprising result is that the optimum can sometimes lie in between the two, neither an ensemble or an E2E system. The work also uncovers links to Dropout, and raises questions around the nature of ensemble diversity and multi-branch networks.