MGPO selects planning computations in approximately 0.03 seconds, over 100 times faster than baseline methods¶
Claim · All claims
Evidence · 1 study · causalr3·q1
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q1 i? MGPO's computational efficiency enables online VOC computation, a prerequisite for real-time feedback in an intelligent tutor. → Heindrich 2025
Evidence¶
Heindrich 2025¶
Heindrich, L., Consul, S., & Lieder, F. (2025). An Intelligent Tutor for Planning in Large Partially Observable Environments. International Journal of Artificial Intelligence in Education, 35, 3095–3127. https://doi.org/10.1007/s40593-025-00493-7
q1 · i? · causal · r3
Timing data from the simulation evaluation section. The article reports "MGPO took 0 .03 seconds" per computation selection, while other baselines required over 100 times the computational budget (meta-greedy 4.89s, PO-UCT 5000 steps 14.19s).
"On average, MGPO took 0 .03 seconds to select the next computation. The only baseline method with a comparable runtime is PO-UCT with 10 steps (0 .02 seconds)."
Discussion¶
Related Claims¶
- MGPO discovers significantly more resource-rational planning strategies than baseline methods in partially observable environments — related
- A preregistered experiment with 330 participants demonstrated the intelligent tutor is highly effective at improving planning in partially observable environments — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
mgpo-computational-efficiency |
title |
MGPO selects planning computations in approximately 0.03 seconds, over 100 times faster than baseline methods |
description |
MGPO selects planning computations in approximately 0.03 seconds, over 100 times faster than baseline methods |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2025 |
generated |
by process:wiki-ingest on 2026-10-28 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — heindrich-2025 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.