The guide for companies on machine learning
“The field of AI is developing at a rapid pace, and hardly any company is (or should be) able to tackle all the issues on its own. A systematic approach to the make-or-buy decision is needed. However, to date most companies have not approached this question systematically at all or, even worse, they have simply delegated this decision to their standard IT purchasing process.”
According to Gartner, most AI projects fail and around 80% never reach production. This guide is an attempt to share the experiences and insights of ML practitioners in the corporate sector in Germany.
Here, the challenges and best practices that companies face in practice are discussed and explained. The guide is managed by appliedAI and reflects the commitment of appliedAI's partners to share their experiences and insights beyond the PoC stage.
This guide consists of four topics:
- The ML lifecycle
- ML architectures
- Platforms for ML
- Challenges and best practices
This report is the result of the appliedAI Enterprise ML working group and is based on the experiences of leading experts from appliedAI partner companies.
Authors of the whitepaper
- Alexander Waldmann, former colleague at appliedAI Initiative
- Alexander Machado, Head of MLOps Processes at appliedAI Initiative
We thank you for your contributions
- Jörn Franke (European Central Bank)
- Faizan Aslam and Simon-Pierre Genot (Infineon)
- Dilek Sezgün (IBM)
- Anant Nawalgaria (Google)
- Dirk Wacker and Into Kemmerzell (Giesecke+Devrient GmbH)
- Matthias Neuehofer (Baywa AG)
- Efrem Ghebru and Michael Ksoll (EnBW)
and many more.
The Enterprise Guide to Machine Learning
According to Gartner, most AI projects fail and about 80% never reach production. This guide is an attempt to share the experiences and learnings of ML practitioners in the enterprise sector in Germany. Here we discuss and elaborate on the challenges and best practices companies face in the real world. Maintained by the appliedAI, the guide reflects a commitment from the partners of appliedAI to share their experiences and learnings when moving beyond the PoC.
This guide consists of four topics:
- The ML lifecycle: This topic focuses on the importance of having a well-defined ML lifecycle. It condenses ML lifecycle models from different technology players including appliedAI.
- ML architectures: This topic focuses on the requirements and abstract solution designs for the outlined lifecycles. The discussion will largely center on a collection of approaches to solving common problems encountered along the ML lifecycle.
- Platforms for ML: This topic finally answers which tools and platforms have been used by our partners to implement particular architectures.
- Challenges and best practices: This topic condenses the technical and non-technical insights we received from our partners and enriched them with appliedAI internal experiences learned over the last years. It includes perspectives on the practitioner's technical challenges and best practices along the ML lifecycle, management of AI projects, AI team working, and platforms usage.
This report is the result of the appliedAI working group “Enterprise ML” and has drawn on the experience of leading experts from appliedAI partner companies.
Thank you for your contributions: Jörn Franke from the European Central Bank, Faizan Aslam and Simon-Pierre Genot from Infineon, Dilek Sezgün from IBM, Anant Nawalgaria from Google, Dirk Wacker and Ingo Kemmerzell from Giesecke+Devrient GmbH, Matthias Neuenhofer from Baywa AG, Efrem Ghebru and Michael Ksoll from EnBW, Hermann Wedlich, Muneer Ahmad and Gerhard Wolf from NetApp and many more.
Authors of the guide include Alexander Waldmann and Alexander Machado.