AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Avoiding Side Effects By Considering Future Tasks

Victoria Krakovna, Laurent Orseau, Richard Ngo, Miljan Martic, Shane Legg

NeurIPS 2020 · Reinforcement learning › Visual and representation RL · approach: Inferring decision-making from behavior · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
Add the baseline policy to the exact future-task reward (ablation)Future task reward · Sushiinterference behaviors per episode: 1 -> 0Improved
Add the baseline policy to the UVFA future-task reward (ablation)Future task reward · Sushiinterference behaviors per episode: 0.64 $\pm$ 0.07 -> 0.05 $\pm$ 0.01Improved
Add the baseline policy to the exact future-task reward (ablation)Future task reward · Vaseside effects per episode: 0 -> 0No clear effect
Add the baseline policy to the exact future-task reward (ablation)Future task reward · Boxside effects per episode: 0 -> 0No clear effect
Add the baseline policy to the exact future-task reward (ablation)Future task reward · Soko-coin-2side effects per episode: 0 -> 0No clear effect
Add the baseline policy to the UVFA future-task reward (ablation)Future task reward · Vaseside effects per episode: 0.12 $\pm$ 0.05 -> 0.12 $\pm$ 0.05No clear effect
Add the baseline policy to the UVFA future-task reward (ablation)Future task reward · Boxside effects per episode: 0.22 $\pm$ 0.06 -> 0.22 $\pm$ 0.06No clear effect
Add the baseline policy to the UVFA future-task reward (ablation)Future task reward · Soko-coin-2side effects per episode: 0.53 $\pm$ 0.09 -> 0.53 $\pm$ 0.09No clear effect
Add the baseline policy to the UVFA future-task reward (ablation)Future task reward · Soko-coin-3side effects per episode: 1.04 $\pm$ 0.12 -> 1.04 $\pm$ 0.12No clear effect
Replace the exact future-task reward with its UVFA approximation (ablation)Future task reward with baseline · Vaseside effects per episode: 0 -> 0.12 $\pm$ 0.05Mixed
Replace the exact future-task reward with its UVFA approximation (ablation)Future task reward with baseline · Boxside effects per episode: 0 -> 0.22 $\pm$ 0.06Mixed
Replace the exact future-task reward with its UVFA approximation (ablation)Future task reward with baseline · Soko-coin-2side effects per episode: 0 -> 0.53 $\pm$ 0.09Mixed

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