Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Pixel-by-pixel Image Classification on MNIST ordered
Loading...
99.4
Accuracy
Lipschitz RNN using Euler
93.888
95.319
96.75
98.181
Jun 22, 2020
Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
Lipschitz RNN using Euler
N=128, # params=≈31K
2020.06
99.4
Lipschitz RNN using RK2
N=128, # params=≈34K
2020.06
99.3
MomentumLSTM
N=256, # params=≈270K
2020.06
99.1
Lipschitz RNN using RK2
N=64, # params=≈9K
2020.06
99.1
Lipschitz RNN using Euler
N=64, # params=≈9K
2020.06
99
Exponential RNN
N=360, # params=≈69K
2020.06
98.4
Incremental RNN
N=128, # params=≈4K/8K
2020.06
98.1
Antisymmteric RNN
N=128, # params=≈10K
2020.06
98
LSTM baseline
N=128, # params=≈68K
2020.06
97.3
Full Capacity Unitary RNN
N=512, # params=≈270K
2020.06
96.9
Kronecker RNN
N=512, # params=≈11K
2020.06
96.4
Unitary RNN
N=512, # params=≈9K
2020.06
95.1
Sequential NAIS-Net
N=128, # params=≈18K
2020.06
94.3
Soft orth. RNN
N=128, # params=≈18K
2020.06
94.1
Feedback
Search any
task
Search any
task