Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Next-step prediction on Synthetic HMM Undercomplete case, N=4000
Loading...
1.368
Validation Loss
HMM Filter
1.22864
2.16932
3.11
4.05068
Nov 13, 2025
Validation Loss
Updated 1mo ago
Evaluation Results
Method
Method
Links
Validation Loss
HMM Filter
2025.11
1.368
nanoGPT-s
# parameters=73,920, t...
2025.11
1.458
nanoGPT-m
# parameters=221,760,...
2025.11
1.475
Belief Net
# parameters=12,352, t...
2025.11
1.569
Spectral
# parameters=524,416,...
2025.11
1.624
Baum-Welch
# parameters=12,352, t...
2025.11
1.951
Random
2025.11
4.852
Feedback
Search any
task
Search any
task