Large-Scale Deep Learning on the YFCC100M Dataset

TL;DR

This paper reports on training a 15-billion-parameter deep neural network on the largest publicly available image and video dataset, demonstrating the feasibility of large-scale unsupervised learning on HPC architectures.

Contribution

It introduces a large-scale deep learning experiment using a 15-billion-parameter network on the YFCC100M dataset, highlighting the potential of scaling up models and datasets.

Findings

Training took eight days on 98 GPUs.

Preliminary results show promise for large-scale unsupervised learning.

Discusses future research directions.

Abstract

We present a work-in-progress snapshot of learning with a 15 billion parameter deep learning network on HPC architectures applied to the largest publicly available natural image and video dataset released to-date. Recent advancements in unsupervised deep neural networks suggest that scaling up such networks in both model and training dataset size can yield significant improvements in the learning of concepts at the highest layers. We train our three-layer deep neural network on the Yahoo! Flickr Creative Commons 100M dataset. The dataset comprises approximately 99.2 million images and 800,000 user-created videos from Yahoo's Flickr image and video sharing platform. Training of our network takes eight days on 98 GPU nodes at the High Performance Computing Center at Lawrence Livermore National Laboratory. Encouraging preliminary results and future research directions are presented and discussed.