List your company
ResearchFormerly WizWand

Image Classification on MNIST (train) (Training Loss)

0Training LossDFC-SS
-0.08720.50141.091.6786Jun 15, 2021Apr 8, 2022Jan 31, 2023Nov 25, 2023Sep 18, 2024Jul 13, 2025May 7, 2026
Updated 5mo ago

Evaluation results

MethodLinks
DFC-SS2021.060
DFC-SSA2021.060
DFC2021.060
BP2021.060
DFA2021.060
Evolved (APR)LR=Fixed, Optimizer=Adam, Epochs=1002026.030
Random Reshuffle (RR)LR=Fixed, Optimizer=Adam, Epochs=1002026.030
Shuffle Once (SO)LR=Fixed, Optimizer=Adam, Epochs=1002026.030
Incremental Gradient (IG)LR=Fixed, Optimizer=Adam, Epochs=1002026.030
DFC (fixed)2021.060
DFC-SS (fixed)2021.060
DFC-SSA (fixed)2021.060
Shuffle Once (SO)LR=Diminishing, Optimizer=Adam, Epochs=1002026.030
Incremental Gradient (IG)LR=Diminishing, Optimizer=Adam, Epochs=1002026.030
Evolved (APR)LR=Diminishing, Optimizer=Adam, Epochs=1002026.030
Random Reshuffle (RR)LR=Diminishing, Optimizer=Adam, Epochs=1002026.030
AdamWweight decay=0.01, learning rate=10^−3, batch size=128, seeds=32026.050.0002
GONOlearning rate=10^−3, batch size=128, seeds=32026.050.0004
GGDNets=CNN_3, Structures=3*(3*3)+128,102026.020.0034
AdamNets=CNN_3, Structures=3*(3*3)+128,102026.020.0075
GGDNets=CNN_2, Structures=3*(3*3)+102026.020.0116
AdamNets=CNN_2, Structures=3*(3*3)+102026.020.0132
GGDNets=CNN_1, Structures=2*(3*3)+102026.020.0136
AdamNets=CNN_1, Structures=2*(3*3)+102026.020.0174
MuonNets=CNN_3, Structures=3*(3*3)+128,102026.020.019
MuonNets=CNN_2, Structures=3*(3*3)+102026.020.0232
MuonNets=CNN_1, Structures=2*(3*3)+102026.020.0394
FldzhyanMesh=784x784, Epochs=150, Batch Size=512, Learning Rate=0.0000382025.120.0588
Fldzhyan BellMesh=784x784, Epochs=150, Batch Size=512, Learning Rate=0.0000382025.120.0634
Adamlearning rate=10^−3, batch size=128, seeds=32026.050.067
SGDMNets=CNN_1, Structures=2*(3*3)+102026.020.0672
SGDMNets=CNN_3, Structures=3*(3*3)+128,102026.020.0805
Clements BellMesh=784x784, Epochs=150, Batch Size=512, Learning Rate=0.0000382025.120.0851
SGDMNets=CNN_2, Structures=3*(3*3)+102026.020.0856
ClementsMesh=784x784, Epochs=150, Batch Size=512, Learning Rate=0.0000382025.120.1026
SGDNets=CNN_1, Structures=2*(3*3)+102026.020.196
SGDNets=CNN_2, Structures=3*(3*3)+102026.020.293
SGDNets=CNN_3, Structures=3*(3*3)+128,102026.020.35
SSGDNets=CNN_1, Structures=2*(3*3)+102026.020.789
SSGDNets=CNN_2, Structures=3*(3*3)+102026.021.91
SSGDNets=CNN_3, Structures=3*(3*3)+128,102026.022.18