AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Grounding Aleatoric Uncertainty for Unsupervised Environment Design

Minqi Jiang, Michael D Dennis, Jack Parker-Holder, Andrei Lupu, Heinrich Kuttler, Edward Grefenstette, Tim Rocktäschel, Jakob Nicolaus Foerster

NeurIPS 2022 · Reinforcement learning › Visual and representation RL · approach: Probabilistic and autonomous RL · 3 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace PLR-perp with SAMPLR (fictitious transitions)PLR-perp · CarRacingF1, q=0.4episodic return: 43 -> 195Improved
replace PLR-perp with SAMPLRPLR-perp · CarRacingF1, q=0.6episodic return: -115 -> -1Improved
replace PLR-perp with SAMPLRPLR-perp · CarRacingF1, q=0.8episodic return: -151 -> -111Improved
replace PLR-perp with SAMPLRPLR-perp · CarRacingF1, q=0.2episodic return: 323 -> 393Improved
replace naive grounding with SAMPLRPLR-perp with naive grounding · CarRacingF1, q=0.4episodic return: 75 -> 195Improved
replace naive grounding with SAMPLRPLR-perp with naive grounding · CarRacingF1, q=0.6episodic return: -79 -> -1Improved
replace naive grounding with SAMPLRPLR-perp with naive grounding · CarRacingF1 ground truth q~Beta(1,15)episodic return: 618 -> 616No clear effect
replace DR with SAMPLRDR · CarRacingF1, q=0.4episodic return: 94.7 -> 195Improved
replace DR with SAMPLRDR · CarRacingF1 ground truthepisodic return: 581 -> 616Improved
use plain PLR-perp curriculum instead of DR (no grounding)DR · CarRacingF1 ground truthepisodic return: 581 -> 543Hurt
add naive grounding to PLR-perpPLR-perp · CarRacingF1 ground truthepisodic return: 543 -> 618Improved
add naive grounding to PLR-perpPLR-perp · Stochastic Fruit ChoicereturnImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy