AI Paper Atlas

Research map / Video / Video segmentation and tracking

Hierarchical Visuomotor Control of Humanoids

Josh Merel, Arun Ahuja, Vu Pham, Saran Tunyasuvunakool, Siqi Liu, Dhruva Tirumala, Nicolas Heess, Greg Wayne

ICLR 2019 · Video › Video segmentation and tracking · approach: Physics-inspired motion prediction · 68 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 discrete selection among control fragments instead of the compared control approachesHumanoid visuomotor control · Go-to-target, Walls, Gaps, Forage, Heterogeneous Foragetask performanceImproved
Use hierarchical control with pretrained low-level skills instead of training a flat policy from scratchHumanoid visuomotor control · Walls, Go-to-targettask performanceImproved
Use hierarchical control with pretrained low-level skills instead of training a flat policy from scratchHumanoid visuomotor control · Foragetask performanceImproved
Use control fragments rather than a fixed-turning-radius steerable controllerHumanoid visuomotor control · Go-to-targettask performanceImproved
Use control fragments rather than a structured switching controller with predefined transitionsHumanoid visuomotor control · Wallsforward velocityImproved
Use control fragments rather than a structured switching controller with predefined transitionsHumanoid visuomotor control · Foragetask performanceImproved
Use a query-based selector instead of discrete control-fragment selectionControl-fragment high-level controller · Core taskstask performanceHurt
Add low-level policy pretraining by supervised imitation before reinforcement learningSingle-clip tracking policies · Motion-capture trackingtraining convergence and resulting policy qualityImproved
Use control fragments longer than one time step (ablation)Control-fragment controller · Go-to-targetearly explorationImproved
Use control fragments longer than one time step (ablation)Control-fragment controller · Go-to-targetultimate task performanceHurt
Add functionally similar, redundant behaviors to the control-fragment set (ablation)Control-fragment controller · Go-to-targettraining speedImproved
Add extraneous policies involving movements unrelated to locomotion (ablation)Control-fragment controller · Go-to-targettraining performanceNo clear effect

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