节选 Measuring personal growth

During a reinforcement learning course in college, I learned about empowerment maximization. It’s a simple principle that enables robots/agents to exhibit relatively intelligent behavior.

In the face of uncertainty, an agent following empowerment maximization would choose the action that maximizes future options. For example, facing multiple switches, it’d choose the switch that opens the most doors.

I realized that this is the same principle that I’ve followed. In the face of uncertainty, I lean towards the decision that would give me the most future options.

For example, I’d choose a job that pays less but gives me more job options in the future (e.g. if the job gives me exposure like allowing me to work on open source or publish papers).

I’d prioritize tasks that teach me transferable skills instead of tasks that teach me niche, narrow skills.

Perhaps I can measure my growth by how many new options I have gained/lost. What options are available to me today that were not available to me 5 years ago? What options were available to me 5 years ago that aren’t available to me now?

More importantly, what options that are not available to me today do I want 5 years from now?

在大学的强化学习课程中，我学习了赋能最大化。 这是一个简单的原理，它能让机器人/代理表现出相对智能的行为。

面对不确定性，遵循授权最大化原则的代理会选择最大化未来选项的行动。例如，面对多个开关，它会选择打开最多门的开关。

我意识到这和我一直遵循的原则是一样的。面对不确定性，我倾向于做出能给我带来最多未来选择的决定。

例如，我会选择一份薪水较低但能给我未来更多工作选择的工作（例如，如果这份工作能给我带来曝光机会，比如允许我从事开源工作或发表论文）。

我会优先考虑那些教我可转移技能的任务，而不是那些教我小众、狭窄技能的任务。

也许我可以用我获得/失去多少新的选择来衡量我的成长。今天我有哪些五年前没有的选择？五年前我有哪些选择，现在却没有了？

更重要的是，我希望在五年后获得哪些今天无法获得的选择？

原文链接：Measuring personal growth