The workshop was fun, informative, inspiring, and thought-provoking. We also enjoyed the lively discussion from all participants.

The key takeaway is that, to exploit machine learning to its full potential, we need to let the machine learn and constantly learn. With reinforcement learning (RL) and multi-arm bandits (MABs), a machine, or algorithm, can learn, predict and act optimally upon every interaction. Classic RL example is self-driving cars, in which the car constantly learns about the environment, predict if the surrounding object is, e.g., another car, a dog, or a person, and drive, and repeat. In marketing, RL can be applied to morph the website into consumer cognitive styles and/or where they are on the purchase funnel. In clinical trials, MABs can potentially save lives (working paper).

If you are curious and would like to learn more about RL/MABs with hands-on examples from real-world data and problems, Gui is teaching a summer course on August 22-26, 2022. Please find more details here.