MZ+Recon

Reported on 3 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology3 results

Meta-LearningonML45
Meta-test success rate (zero-shot)· 2021-11-02
18.5
SOTA
Procedural Generalization by Planning with Self-Supervised World Models arXiv:2111.01587
Meta-LearningonML45
Meta-train success rate· 2021-11-02
74.9
best: 77.2 (MZ)
Procedural Generalization by Planning with Self-Supervised World Models arXiv:2111.01587
Meta-LearningonML10
Meta-test success rate (zero-shot)· 2021-11-02
25
best: 26.5 (MZ)
Procedural Generalization by Planning with Self-Supervised World Models arXiv:2111.01587