Transcript
Now that we’ve touched on AI, let’s move on to machine learning. Machine learning really is a subset of AI, and often when people refer to AI, they are usually talking about machine learning, especially when it comes to FinTech.
For example, advances in algorithmic trading are being powered by machine learning. Additionally, the ability of financial institutions to manage risk, detect fraud, and optimize operational processes are all being made more efficient and accurate through machine learning.
Even lawyers like us, – former lawyers, who maybe thought we were immune from technological change are being impacted as machine learning technology is already being implemented to review documents, like contracts or loan agreements much faster, cheaper, and more accurately than a human could.
Sounds exciting right? So what is machine learning?
Machine learning is effectively a machine, say a computer, combing through and analyzing with statistics large amounts of data to find patterns. That data could be in the form of text, like in a loan document, or it could be a series of numbers, like stock prices, or a whole host of other types of information. Based on that data, the machine can start making predictions and as more data comes in, the predictions become more refined. Most of us interact with machine learning almost daily—basically whenever enjoying any service that recommends things to us, like new shows that Netflix recommends to you tonight.
Lastly, machine learning can be further specified as supervised learning, where the data is labelled or identified; unsupervised learning, where there are no such identifying markers; or reinforcement learning, which is what Google’s AlphaGo represents, and based on the machine figuring things out after exploring multiple permutations of outcomes—basically a massive iterative process of trial and error.
