notesum.ai

Published at December 4

Experience-driven discovery of planning strategies

cs.AI

Released Date: December 4, 2024

Authors: Ruiqi He¹, Falk Lieder²

Aff.: ¹Max Planck Institute for Intelligent Systems Tuebingen, Germany; ²UCLA Department of Psychology Los Angeles, United States

Arxiv: http://arxiv.org/pdf/2412.03111v1

Variable	Coefficient	std	t-value	p-value	0.025	0.975
Participant	0.074	0.003	25.872	<.001	0.068	0.080
Habitual model	0.037	0.004	9.103	<.001	0.029	0.045
MF Reinforce	0.040	0.004	9.920	<.001	0.032	0.048
Non-learning	0.107	0.004	26.415	<.001	0.099	0.115
SSL	-0.034	0.004	-8.499	<.001	-0.042	-0.026
Hybrid Reinforce	0.034	0.004	8.304	<.001	0.026	0.042
Participant : Trial	0.002	$4.11e^{-5}$	51.242	<.001	0.002	0.002
Habitual model : trial	-0.001	$5.81e^{-5}$	-21.258	<.001	-0.001	-0.001
MF Reinforce : Trial	-0.002	$5.81e^{-5}$	-26.658	<.001	-0.002	-0.001
Non-learning : Trial	-0.002	$5.81e^{-5}$	-36.189	<.001	-0.002	-0.002
SSL : Trial	-0.002	$5.81e^{-5}$	-31.573	<.001	-0.002	-0.002
Hybrid Reinforce : Trial	-0.002	$5.81e^{-5}$	-26.184	<.001	-0.002	-0.001