Report on Praise data text-classification in pursuit on an Actor-Critic RL agent for Headless DAO steering

Provide as much information as possible about your idea.

I went overboard analyzing the Praise Quant data this month as a practice problem/skill development exercise for Reinforcement Learning.

How is it relevant to the TEC mission, vision and values?

I am a bit out of the loop regarding TEC’s is currently at with the rewards system. I just wanted to share the report to see what peoples’ impressions are.

I like working with written language and have been working towards a similar control-system/feedback loop for a power-electronics project.

What can I say, to a man training up in ‘Ostrom-style IAD’, every problem looks like a ‘Markov Decision Process’ :yum:

Who is going to be affected by what you’re proposing?

Im not really proposing anything at this time, just sharing the report to see what kind of impressions it gets.

Who has expertise in this topic and could advise your process?

Rewards Team? I helped Krisofer with some of the data parsing earlier this year.

What type of proposal is this?

  1. This is an idea or something I just need advice to proceed with.