Re:Show HN: Factorio Learning Environment – Agents Build Fac...
by noddybear on Tuesday March 11, 2025 @12:47PM
(#43334396)
I think generating the scenarios as you suggest (in text) is easy, but creating correct factory game states to start from is a lot harder. AFAIK it reduces into the same manual task of designing an init state and a task to complete.
Re:Show HN: Factorio Learning Environment – Agents Build Fac...
by infogulch on Tuesday March 11, 2025 @12:59PM
(#43334572)
Yes each scenario will need someone to design it, but you can get a lot of mileage out of each. E.g. consider the "place the missing power pole" scenario: manually build a factory with a few dozen machines connected to a couple steam engines with 20 power poles, then you can generate 400 playable puzzles/scenarios by deleting 1-2 power poles from the working starting point. Humans would find all of these to be equivalent, but I think agents need the explicit variation to learn the lesson properly.
Re:Show HN: Factorio Learning Environment – Agents Build Fac...
by noddybear on Wednesday March 12, 2025 @09:32AM
(#43343072)
Oh super interesting! Create 10 scenarios containing working factories, and ‘drop out’ entities to break the factory in different ways. great idea.
Re:Show HN: Factorio Learning Environment – Agents Build Fac...
by infogulch on Wednesday March 12, 2025 @03:03PM
(#43346582)
Yes exactly! This approach can generate hundreds of "fix the problem"-type tests very easily. With some creative thinking I suspect you can use variations to stack multipliers on other types of tests as well.
Re:Show HN: Factorio Learning Environment – Agents Build Fac...
by martbakler on Tuesday March 11, 2025 @01:14PM
(#43334754)
We are thinking of something like this (a curriculum approach) for further training. The reason why we didn't want to do this for current work, where the emphasis is on evaluations, is that the "difficulty level" of different tasks is quite subjective and hence we would need to make arbitrary decisions that could affect the evals (i.e which tasks would follow which scenarios, how to ensure sufficient coverage across all difficulty levels etc)
Re:Show HN: Factorio Learning Environment – Agents Build Fac...
by infogulch on Tuesday March 11, 2025 @02:04PM
(#43335321)
"a curriculum approach" is a nice way to put it!
> the difficulty level of different tasks is subjective
That makes sense. I wonder if difficulty of different scenarios could be derived by assuming a partial ordering and ranking based on training rate: e.g. it preforms better at scenario T if it trains scenario A first, but training scenario first B doesn't help with T. Then infer A < T, and B ? T.
I hypothesize that ML agents learn faster when evaluated on a sample from a large bank of scenarios of smoothly increasing complexity where more complex scenarios are presented after it scores sufficiently high on lower complexity scenarios.