Cultivating DNN Diversity for Large Scale Video Labelling
This addresses the challenge of improving ensemble performance in video understanding for researchers and practitioners, but it is incremental as it builds on known ensemble benefits.
The paper tackled the problem of promoting and measuring DNN diversity for large-scale video labeling, showing that diversity can be cultivated through unexpected means like model over-fitting or dropout variations, and their solution ranked #7 in the Kaggle competition.
We investigate factors controlling DNN diversity in the context of the Google Cloud and YouTube-8M Video Understanding Challenge. While it is well-known that ensemble methods improve prediction performance, and that combining accurate but diverse predictors helps, there is little knowledge on how to best promote & measure DNN diversity. We show that diversity can be cultivated by some unexpected means, such as model over-fitting or dropout variations. We also present details of our solution to the video understanding problem, which ranked #7 in the Kaggle competition (competing as the Yeti team).