AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

SMILe: Scalable Meta Inverse Reinforcement Learning through Context-Conditional Policies

Seyed Kamyar Seyed Ghasemipour, Shixiang (Shane) Gu, Richard Zemel

NeurIPS 2019 · Reinforcement learning › Imitation learning · approach: Theory of inverse RL · 19 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
use max-ent IRL (SMILe state-action) instead of behavioral cloning (Meta-BC)context-conditional meta-imitation policy · HalfCheetah Random Velocity, 4 demos 20 subabs velocity error (lower better): 0.931 -> 0.181Improved
use max-ent IRL (SMILe state-action) instead of behavioral cloning (Meta-BC)context-conditional meta-imitation policy · HalfCheetah Random Velocity, 64 demos 20 subabs velocity error (lower better): 0.468 -> 0.102Improved
use max-ent IRL (SMILe state-action) instead of behavioral cloning (Meta-BC)context-conditional meta-imitation policy · Walker random dynamicscustom normalized metric: -0.403 -> 0.710Improved
use SMILe instead of Meta-DAgger (online expert BC)context-conditional meta-imitation policy · Walker random dynamicscustom normalized metric: 0.275 -> 0.710Improved
use SMILe instead of Meta-DAgger (online expert BC)context-conditional meta-imitation policy · HalfCheetah Random Velocity, 64 demos 1 subabs velocity error (lower better): 0.082 -> 0.118Hurt
use SMILe instead of Meta-BCcontext-conditional meta-imitation policy · Ant Linear Classificationsuccess rate: 0.50 -> 0.68Improved
use state-only demonstrations instead of state-action (ablation)SMILe · HalfCheetah Random Velocity, 64 demos 1 subabs velocity error (lower better): 0.118 -> 0.138No clear effect
increase number of expert demonstrations per task from 4 to 64 (ablation)Meta-BC · HalfCheetah Random Velocityabs velocity error (lower better): 0.931 -> 0.468Improved

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