femi ...one thousand years later

existential dread

It’s all because of cookie clicker.

Recently, I’ve been playing it a lot more and a friend brought up the idea of trying to mathematically optimize the game. Naturally, the first thing I did was head to google scholar and search for cookie clicker research papers.

I found this pretty great Masters paper from MIT of all places, and wondered just what kind of legend spends two years to write a paper on optimizing a (very well made) idle game?

Kai Xiao finished his undergrad in Mathematics and Computer Science and Ph.D in Computer Science (with a focus on robustness in machine learning) at MIT. During this time he interned at multiple quantitative trading firms, but he now works at OpenAI as a Technical Lead.

This guy has lived a life damn near identical to the life I was planning to lead. Yet the end of his road doesn’t seem that appealing. The worst part is that I can’t quite put my finger on what bothers me about this. I’ve stalked Kai pretty extensively, and to me he seems like he’s living his best life, but is that my best life?

Where do I see my self in 10, 20, 40, or 80 years?

in search of an end goal

So what’s my meaning of life to me? Fame, Fortune, Legacy?

I don’t think either of these three will really satisfy me in the end.

legacy

The closest thing we can have to immortality is a legacy. If we play our cards right, our impact might far outlive us.

I want to leave an impact on the world before I die, but what does this impact matter when I’m dead? I don’t want to do it for massive amounts of fame or money, some deeper need calls me to do it.

I’ve thought about this a lot this week, and to chase this idea of “legacy”, and “maximizing my impact” seems kind of silly. I rarely think about the people who you might regard as having the largest legacies (Steve Jobs, Issac Newton, MLK), but I do think a lot about the people who’ve astonished me with their breadth and depth of knowledge. I want to be more like that. Leaving an impact by how I do things, not what exactly I did.

Not everyone might remember me or the things I did, but in the limit, we’ll all be skeletons anyway. I won’t live my life just so that Historians of Far Away Days deem me worthy of jotting down.

where to from here

So the answer is, just work on things I’m passionate about? But what am I really passionate about. New fields rise and fall, what do I want to spend my life doing?

I don’t know! But what I do now is that I wouldn’t let a version of me from 5 years ago control me right now, hell, I wouldn’t let me 2 weeks ago control what I eat for breakfast. The same will probably hold true in the future.

Yet, my actions now have a direct impact on the options available to me in the future. I may not know where I want to go now, but I’ll have more information later. So maybe the optimization target is “what can I do now that’ll give me the most options in the future.” (Paul Graham calls this Staying upwind)

goals for 2024: the grand finale

I know what you’re thinking. “Here we go, femi once again is moving the goal posts.” But I feel shameless about constantly revising my goals like this. I’m always learning more about the world, and what I want to do. To be clear, I’m fine with revising longer term goals. I often fail to completely achieve my shorter term goals (\(\le\) 1 week), which I don’t think I should revise at all.

Goal for 2024: Learn a lot about a lot

Sub goals:

  1. Always be reading a new book/course on some generally useful topic.
  2. Always be working on some generally useful project.
  3. Maintain good grades (~3.8 GPA)
  4. Read a research paper daily
  5. Having fun - not just being a CS robot

I’ll be focusing on broader projects and less on specific career paths. This means less of a focus on research (which has to be specific and targeted) and more of a focus just general learning (where I’m free to learn and explore).

Weekly goals:

  1. Finish 2 Chapters of Understanding Deep Learning
  2. Redesign UI for engineering project
  3. Don’t fail tests lol.
  4. Write a blog post about continual learning
  5. Read two chapters of TSCM

You’ll note that the bar here is signifcantly lower than other weeks. I’m accounting for RANDOM SHIT which seems to be a constant in my life. We’ll see if I can get it done. As always, I’ll let you know how it goes.