Challenges for Machine Learning Systems toward Continuous Improvement

Abstract

When executing machine learning pipelines for trainings and inferences, the systems and machine learning infrastructures vary depending on required characteristics and requirements such as the purpose of the application, data volume, and latency. On the other hand, many companies in industry have built machine learning infrastructures with each companies knowledges. The knoledges are not organized yet since they are engineering efforts and engineers less motivated to publish them so that there are few papers for the system design and problems characteristic in machine learning infrastructure and systems. In this presentation, we will introduce challenges for machine learning systems, especially for continuous prediction in the production environment and approaches to them.

Date
2019-11-22 14:20 — 15:00
Location
Winc Aichi, Nagoya
Avatar
Aki Ariga
Machine Learning Engineer

Interested in Machine Learning, ML Ops, and Data driven business.

Related