AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

Robust Imitation via Mirror Descent Inverse Reinforcement Learning

Dong-Sig Han, Hyunseo Kim, Hyundo Lee, JeHwan Ryu, Byoung-Tak Zhang

NeurIPS 2022 · Reinforcement learning › Imitation learning · approach: Reward learning from demonstrations · 1 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
replace RAIRL with MD-AIRL (Shannon)RAIRL · bandit |A|=10^3Bregman divergence x |A|: 140.3 -> 125.3Improved
replace RAIRL with MD-AIRL (Tsallis)RAIRL · bandit |A|=10^3Bregman divergence x |A|: 0.55 -> 0.24Improved
replace RAIRL with MD-AIRL (Tsallis)RAIRL · bandit |A|=10^4Bregman divergence x |A|: 4.95 -> 4.21Improved
replace RAIRL with MD-AIRL (sin)RAIRL · bandit |A|=10^3Bregman divergence x |A|: 3.82 -> 1.07Improved
replace RAIRL with MD-AIRL (Shannon)RAIRL · bandit |A|=10^2Bregman divergence x |A|: 2.55 -> 2.28Improved
replace RAIRL with MD-AIRL (exp)RAIRL · bandit |A|=10^4Bregman divergence x |A|: 5.06 -> 4.97No clear effect
replace RAIRL with MD-AIRL (Tsallis)RAIRL · Walker2d noisy demos eps=0.01score: 2731.84 -> 3624.00Improved
replace RAIRL with MD-AIRL (Shannon)RAIRL · Walker2d noisy demos eps=0.5score: 2451.00 -> 3252.65Improved
replace RAIRL with MD-AIRL (Shannon)RAIRL · Ant noisy demos eps=0.5score: 3777.78 -> 4284.38Improved
replace RAIRL with MD-AIRL (Tsallis)RAIRL · HalfCheetah noisy demos eps=0.5score: 4216.67 -> 4247.44Improved
replace RAIRL with MD-AIRL (Shannon)RAIRL · Hopper noisy demos eps=0.01score: 3636.03 -> 3669.25No clear effect
replace RAIRL with MD-AIRLRAIRL · MuJoCo, 32 configurationsaverage score: 30 of 32Improved

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