What is ML governance and why is it needed?

ML governance is a set of policies, processes, and activities by which an organization manages, controls, and monitors an ML model's life cycle, dependencies, access, and performance to avoid or minimize financial risk, reputation risk, compliance risk, and legal risk.

The stakes in model risk management are high. To put this into context, let's revisit the impact of the financial crisis in 2007 and 2008 due to inadequate ML governance. Many of us probably still vividly remember the aftermath of the great recession caused by the crisis, where millions of people were impacted in terms of their jobs, investments, or both, and many of the largest financial institutions were brought to their knees and went out of business. The government had to step in to bail out many institutions such as Fannie Mae and Freddie Mac. This crisis was caused in large part by the flawed model risk management process and governance across financial...