Vertex AI, Google Cloud’s controlled machine learning platform, has been announced. Developers should be able to submit and manage their artificial intelligence (AI) models more easily with the platform.
Vertex AI unifies the Google Cloud resources for developing machine learning models into a single UI and API, making the process of creating, training, and deploying machine learning models at scale much easier. Users can transfer models from research to production faster, uncover trends and anomalies more quickly, make better forecasts and decisions, and be more resilient in the face of changing market conditions in this single environment.
In comparison to competing platforms, Google claims that its Vertex AI platform needs approximately 80% fewer lines of code to train a model, allowing data scientists and ML engineers of all levels of expertise to incorporate Machine Learning Operations (MLOps) to effectively design and manage ML projects during the development lifecycle.
“We had two guiding lights while building Vertex AI: get data scientists and engineers out of the orchestration weeds, and create an industry-wide shift that would make everyone get serious about moving AI out of pilot purgatory and into full-scale production,” said Andrew Moore, vice president and general manager of Cloud AI and Industry Solutions at Google Cloud. “We are very proud of what we came up with in this platform, as it enables serious deployments for a new generation of AI that will empower data scientists and engineers to do fulfilling and creative work.”
Vizier, Feature Store
Vertex AI will be integrated with Vizier, Google’s AI optimizer, allowing developers to adapt hyperparameters to machine learning models automatically. This would cut the time it takes to set up a model in half, allowing engineers to run more experiments in less time.
Vertex also has a Feature Store where users can share and reuse Vertex experiments and machine learning features. This would enable them to put their models into action much quicker.