Fully deterministic runs

[jadkins99]

Awesome repo. quick question,

I ran the DMC WalkerWalk experiment 3 different times with the same seeds and got 3 different learning curves. How can I get reproducible experiments?Awesome repo. quick question,

I ran the DMC WalkerWalk experiment 3 different times with the same seeds and got 3 different learning curves. How can I get reproducible experiments?

[danijar]

Hi, are you asking for fully deterministic runs? I haven't paid much attention to this but I think the agent is already fully deterministic, so you'd probably just have to set the environment seed (make sure if you use more than 1 environment instance, that the environments have different seeds so they produce different data).

[jadkins99]

Okay I will try that. Thank you for the quick response! What exactly is an "environment instance"? I couldn't find a clear definition in the paper.

[jadkins99]

Also, how many seeds were the non-Minecraft experiments run for?

[subho406]

+1 on the question above. Maybe it's not that apparent in the paper, could you also provide some clarification on what the confidence intervals denote in the non-minecraft experiments (DMLab, DMC Proprio, Crafter, etc)? Is it std-error across multiple seeds, or std-error across a window of timesteps with a single seed, or something else?

[danijar]

It's mean/std across seeds and at least 3 seeds per task, often more.

[jadkins99]

Update: I seeded dmc_control here. And still got non-deterministic runs. Are there other non-environment sources of randomness not seeded?

[jadkins99]

I found some non-seeded randomness in the repo. Namely here and here. Wouldn't these affect the agent?

[danijar]

I don't think those two methods are run ever. Could you check e.g. by adding asdf to the two methods to see if it errors?

[swannercjj]

Seeding this it removes randomness from the first 1000 steps, but runs are non-deterministic afterwards.