AIoT DevOps and Infrastructure: Difference between revisions

Revision as of 21:15, 1 September 2020

Many new concepts create challenges for AI DevOps

New roles: data scientist, AI engineer
New artefacts (in addition to code): Data, Models
New methods / processes: AI/data-centric, e.g. „Agile CRISP-DM“, Cognitive Project Management for AI (CPMAI)
New AI tools + infrastructure

Additional AI DevOps challenges

Reproduceability of models
Model validation
Versioning: Models, code, data
Lineage: Track evolution of models over time
Testing and test automation: AI requires new methods and infrastructure
Security: Deliberately skewed models as new attack vector / adversarial attacks
Monitoring and re-training: Model decay requires constant monitoring and re-training

@@ Line 20: / Line 20: @@
 [[File:2.3-DevOps-Enterprise.png|800px|frameless|center|Agile DevOps for Cloud and Enterprise Applications]]
 === Agile DevOps for AI ===
+Many new concepts create challenges for AI DevOps
+* New roles: data scientist, AI engineer
+* New artefacts (in addition to code): Data, Models
+* New methods / processes: AI/data-centric, e.g. „Agile CRISP-DM“, Cognitive Project Management for AI (CPMAI)
+* New AI tools + infrastructure
+Additional AI DevOps challenges
+* Reproduceability of models
+* Model validation
+* Versioning: Models, code, data
+* Lineage: Track evolution of models over time
+* Testing and test automation: AI requires new methods and infrastructure
+* Security: Deliberately skewed models as new attack vector / adversarial attacks
+* Monitoring and re-training: Model decay requires constant monitoring and re-training
 [[File:2.3-DevOps-AI.png|800px|frameless|center|DevOps for AI]]
 === Agile DevOps for IoT ===
 [[File:2.3-DevOps-IoT.png|800px|frameless|center|DevOps for IoT]]
 === Agile DevOps for AIoT ===