notesum.ai

Published at December 5

Reinforcement Learning from Wild Animal Videos

cs.RO

cs.CV

cs.LG

Released Date: December 5, 2024

Authors: Elliot Chane-Sane¹, Constant Roux, Olivier Stasse, Nicolas Mansard

Aff.: ¹LAAS-CNRS, Universite de Toulouse, CNRS, Toulouse, France

Arxiv: http://arxiv.org/pdf/2412.04273v1

Refer to caption

Method	Keeping still		Walking		Running		Jumping
Method	$\|vel_{xy}\|\downarrow$	Style	$vel_{x}\uparrow$	Style	$vel_{x}\uparrow$	Style	$\Delta z\uparrow$	Style
RLWAV	12.3 (1.1)	1.0	23.0 (15.4)	0.88	35.1 (12.9)	1.0	18.5 (3.5)	0.88
no soup	14.2 (1.2)	1.0	21.8 (13.9)	0.62	26.0 (17.9)	0.75	10.6 (8.1)	0.38
no curating	8.9 (2.3)	1.0	-0.1 (1.4)	0.0	-0.9 (5.7)	0.12	19.1 (2.0)	0.75