Joint Machine Learning Research Projects in a Unified Cloud Setting

by

in

1. Bank Rakyat Indonesia (BRI) focuses on small-to-medium businesses and microfinance in Indonesia and Southeast Asia.
2. BRI’s Digital BRIBRAIN department develops AI solutions for customer engagement, credit underwriting, anti-fraud, risk analytics, and smart services.
3. BRI partnered with Google Cloud to research fairness analysis in credit scoring, interpret ML model decisions, and perform sentiment analysis of financial chatbots using Vertex AI technology.

Bank Rakyat Indonesia (BRI) is a large bank in Indonesia and Southeast Asia that focuses on small-to-medium businesses and microfinance. BRI has established a Digital Banking Development and Operation Division to implement digital banking and digitalization. One department within this division, Digital BRIBRAIN, develops AI solutions for customer engagement, credit underwriting, anti-fraud and risk analytics, and smart services and operations for business and operational teams.

The AI research team within Digital BRIBRAIN works on projects like the BRIBRAIN Academy, which collaborates with higher education institutions to nurture AI and ML in banking and finance. By partnering with universities, students can study the application of AI in the financial sector, covering topics such as fairness analysis in credit scoring, explainable AI, graph ML, federated learning, and more.

BRI selected Google Cloud products and services, including Vertex AI technology, to provide a sandbox environment for research collaborations with partner universities. This research covers various use cases, including fairness analysis on credit scoring in banking, interpreting ML model decisions for credit scoring using explainable AI, and sentiment analysis of financial chatbots using graph ML.

Google Cloud’s infrastructure and services, such as BigQuery and Vertex AI, met the needs of these projects, enabling BRI to deploy data warehousing, ML, and access management tools efficiently. The use of Vertex AI Workbench allowed researchers to create and manage virtual machine instances for data preparation, model training, and model evaluation. Fine-grained control and management of access to resources were ensured through Identity and Access Management (IAM).

Source link