AI Paper Atlas

Research map / Reinforcement learning / Offline RL

First-Person Activity Forecasting from Video with Online Inverse Reinforcement Learning

Nicholas Rhinehart, Kris Kitani

TPAMI 2018 · Reinforcement learning › Offline RL · approach: Diffusion models for offline decision making · 30 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 stop-based instead of scene-based goal discovery (ablation)DARKO · Home 1 (action labels)mean probability of true goal: 0.438 -> 0.614Improved
use stop-based instead of scene-based goal discovery (ablation)DARKO · Office 2 (action labels)mean probability of true goal: 0.238 -> 0.625Improved
use stop-based instead of scene-based goal discovery (ablation)DARKO · Lab 1 (action labels)mean probability of true goal: 0.426 -> 0.709Improved
replace visual detections with ground-truth labelled detectionsDARKO · goal forecasting, all environmentsmean probability of true goal: mean absolute improvement of 0.27Improved
add action features to state features in reward (ablation)DARKO (labelled detections) · Home 1mean probability of true goal: 0.735 -> 0.851Improved
add action features to state features in reward (ablation)DARKO (labelled detections) · Office 1mean probability of true goal: 0.581 -> 0.700Improved
add action features to state features in reward (ablation)DARKO (labelled detections) · Lab 1mean probability of true goal: 0.892 -> 0.880Hurt
add state and action features over position-only reward (ablation)DARKO (labelled detections) · Home 1mean probability of true goal: 0.674 -> 0.851Improved
add state and action features over position-only reward (ablation)DARKO (labelled detections) · Lab 1mean probability of true goal: 0.886 -> 0.880Hurt
incorporate goal detection uncertainty into value iteration (ablation)DARKO · synthetic noisy goal detectionsaverage goal forecasting probability: p<10^-7Improved
use DARKO online IRL instead of RNN classifiergoal forecasting (visual detections) · Home 1mean probability of true goal: 0.291 -> 0.524Improved
use DARKO instead of Logistic regressiongoal forecasting (labelled detections) · Home 1mean probability of true goal: 0.517 -> 0.851Improved

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