Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Offline Goal-Conditioned Reinforcement Learning on puzzle 4x5
Loading...
9,600
Success Rate
DQC
-384
2,208
4,800
7,392
Dec 11, 2025
Success Rate
Updated 4d ago
Evaluation Results
Method
Method
Links
Success Rate
DQC
Method=Decoupled Q-chu...
2025.12
9,600
NS
Backup=n-step return b...
2025.12
9,300
DQC-naïve
Execution=partial acti...
2025.12
3,300
IQL
2025.12
2,000
QC
Strategy=Q-chunking
2025.12
2,000
OS
Backup=1-step TD-backup
2025.12
1,900
SHARSA
2025.12
100
FBC
2025.12
0
HFBC
2025.12
0
HIQL
2025.12
0
Feedback
Search any
task
Search any
task