Aspire to Potenials for Learners learning framework

I was reading the "Cat Paper" (1) and it has the line: "Self-taught learning framework". I've spent a lot of time learning on my own, trying to refresh my memory on old things or to try new things. So, that line sent me down a side quest (aka. rabbit hole). Anyway, I was interested in whether this framework existed for DIY / self-directed learning from free or inexpensive resources.

It turns out that the authors actually meant a learning model teaching itself on unlabeled data. My curiosity not sated, I followed my curiosity and found a paper entitled "Fosting Self-Direct Growth with Generative AI: Toward a New Learning Analytics Framework" -- what a mouthful! I actually tore through this paper, it's both engaging and approachable... something I look for in these side quests.

TL;DR

My biggest takeaways? A couple of good pointers for building a DIY learning program as well as some language to define my mental models.

• Limit the scope of your "focus".
• 6 and 12 weeks are key milestones for habit forming.
• You need to build a bespoke rubric per program
• There are existing grading scales to measure your progress (i.e. learning analystics frameworks)

Finally, it was a useful read for building more robust mental models for guarding against over-reliance on the LLMs themselves.

1 - "Cat Paper" - Building High-level Features Using Large Scale Unsupervised Learning - Le et. all. 2 - Fostering Self-Directed Growth with Generative AI: Toward a New Learning Analytics Framework - Qianrun Mao