View a markdown version of this page

An intelligent platform approach - Accenture Enterprise AI – Scaling Machine Learning and Deep Learning Models

This whitepaper is for historical reference only. Some content might be outdated and some links might not be available.

An intelligent platform approach

Deep learning is set to achieve transformation across industries and create new opportunities on a scale not seen since the industrial revolution in the 19th century. It comes with great promise, but with it are significant challenges. Most DL models are built on richer algorithmic primitives, and therefore require higher reuse between tasks, rather than training a model from scratch each time there is a new problem at hand, or a new dataset.

For businesses to derive value, ML and DL models need to be productionized, run at scale, and reused across organizations. Lack of scalability, repeatability, and manual processes diminish any value that would be otherwise realized from these powerful models.

A diagram that shows a complete solution for scaling and productionizing DL models with automated pipelines .

Complete solution for scaling and productionizing DL models with automated pipelines

This proposed architecture in the previous diagram is designed to help achieve three goals:

  • Greater, systematic reuse of features and architectures

  • Reduces manual processes

  • Increases speed to market