Sequential Image Classification On Sequential

평가 지표

Permuted Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Permuted Accuracy	Paper Title	Repository
UnICORNN	98.4	UnICORNN: A recurrent model for learning very long time dependencies	-
CKCNN (1M)	98.54%	CKConv: Continuous Kernel Convolution For Sequential Data	-
Dilated GRU	94.6%	Dilated Recurrent Neural Networks	-
FlexTCN-6	-	FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizes	-
LSSL	98.76%	Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layers	-
LEM	96.6%	Long Expressive Memory for Sequence Modeling	-
STAR	-	Gating Revisited: Deep Multi-layer RNNs That Can Be Trained	-
GAM-RHN-1	96.8%	Recurrent Highway Networks with Grouped Auxiliary Memory
CKCNN (100k)	98%	CKConv: Continuous Kernel Convolution For Sequential Data	-
BN LSTM	95.4%	Recurrent Batch Normalization	-
Sparse Combo Net	96.94	RNNs of RNNs: Recursive Construction of Stable Assemblies of Recurrent Neural Networks	-
Temporal Convolutional Network	97.2%	An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling	-
EGRU	95.1%	Efficient recurrent architectures through activity sparsity and sparse back-propagation through time	-
FlexTCN-4	98.72%	FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizes	-
HiPPO-LegS	98.3%	HiPPO: Recurrent Memory with Optimal Polynomial Projections	-
Dense IndRNN	97.2%	Deep Independently Recurrent Neural Network (IndRNN)	-
coRNN	97.34%	Coupled Oscillatory Recurrent Neural Network (coRNN): An accurate and (gradient) stable architecture for learning long time dependencies	-
Adaptive-saturated RNN	96.96%	Adaptive-saturated RNN: Remember more with less instability
LMU	97.2%	Legendre Memory Units: Continuous-Time Representation in Recurrent Neural Networks
ODE-LSTM	97.83%	Learning Long-Term Dependencies in Irregularly-Sampled Time Series	-

0 of 30 row(s) selected.