The advent of “auto-ML” — software that provides methods and processes for creating machine learning code — has led to calls to “democratize” data science and AI.
What is AutoML?
AutoML is short for "Automatic Machine Learning". It's like having a computer program that can make decisions on how to build a machine learning model, instead of having a human do it.
Let's say you went to a charity shop and picked up a bunch of Lego to build the a LEGO castle but you do not have an instruction sheet. You could try building it yourself, but it might take a long time, and you might not be able to build the best castle possible. With AutoML, it's like having a computer program that can try out different ways of building the castle, just like how you might build different versions of the castle. The computer program can build many castles quickly, test them, and then choose the best one to use. This way, you can build the best LEGO castle ever, and you'll learn what works best for building LEGO castles. Having said that there is an Auto ML app for Legos would recommend you to try out.
The idea of enabling data science and AI more accessible to non-data scientists community (including technologists who are not data scientists) can make a lot of business sense as it would accelerate problem solving as they business data is better understood by the business community / nerdy user.
Balancing the Benefits and Challenges
Auto-ML could accelerate building proof of value for the business problem in question. The platform would not articulate challenges around biases around the data set (subsampling / oversampling, framing biases or labelling biases). the model is built with inputs that aren’t available at run time, or the model overfits or underfits the data, or the model was tested against the wrong benchmark
With the raise of "Responsible AI" to ensure proper governance and risk mitigation strategies while adopting AI use case implementation to abide regulatory policies, raising the awareness and educating the citizen development community is key for the successful adoption.
Neglect burnout Syndrome - With lack of structure in awareness and governance can result in AI project not to move from pro-typing phase to production. AI novices spend time developing AI can lead to wasted efforts and internal resources on projects better left on the cutting room floor. Neglect burnout occurs when you aren’t given enough structure, direction, or guidance with the AI project management.
Building Shared Vision
It does not matter if the AI solution solves a business problem for internal or external facing process to an organization. Educating the responsible of use of technology is key to ensure the risk associated is uncovered and mitigated
Book Recommendation
https://www.reidblackman.com/ethical-machines/
Fun Fact
- Auto ML for Legos (https://apps.apple.com/us/app/brickit-app/id1477221636)