Sequential Image Classification on Sequential MNIST

30개 결과 · ⬇ CSV · JSON

Permuted Accuracy

Rank	Model	Permuted Accuracy	Unpermuted Accuracy	Paper	Code	Year
1	SMPConv	99.10	99.75	SMPConv: Self-moving Point Representations for Continuous Convolution	sangnekim/smpconv	2023
2	LSSL	98.76%	99.53%	Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layers	hazyresearch/state-spaces · ag1988/dss	2021
3	FlexTCN-4	98.72%	–	FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizes	rjbruin/flexconv	2021
4	S4	98.70%	99.63%	Efficiently Modeling Long Sequences with Structured State Spaces	state-spaces/s4 · srush/annotated-s4 · ag1988/dss · +5	2021
5	CKCNN (1M)	98.54%	99.32%	CKConv: Continuous Kernel Convolution For Sequential Data	dwromero/ckconv	2021
6	Modified LMU (165k)	98.49%	–	Parallelizing Legendre Memory Unit Training	hrshtv/pytorch-lmu · elisejiuqizhang/jax-lmu	2021
7	UnICORNN	98.4	–	UnICORNN: A recurrent model for learning very long time dependencies	tk-rusch/unicornn	2021
8	HiPPO-LegS	98.3%	–	HiPPO: Recurrent Memory with Optimal Polynomial Projections	HazyResearch/hippo-code · ag1988/dss	2020
9	CKCNN (100k)	98%	99.31%	CKConv: Continuous Kernel Convolution For Sequential Data	dwromero/ckconv	2021
10	ODE-LSTM	97.83%	–	Learning Long-Term Dependencies in Irregularly-Sampled Time Series	mlech26l/learning-long-term-irregular-ts · mlech26l/ode-lstms	2020
11	coRNN	97.34%	99.4%	Coupled Oscillatory Recurrent Neural Network (coRNN): An accurate and (gradient) stable architecture for learning long time dependencies	tk-rusch/coRNN	2020
12	Dense IndRNN	97.2%	99.48%	Deep Independently Recurrent Neural Network (IndRNN)	Sunnydreamrain/IndRNN_pytorch	2019
12	Temporal Convolutional Network	97.2%	99.0%	An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling	timeseriesAI/tsai · locuslab/TCN · philipperemy/keras-tcn · +32	2018
12	LMU	97.2%	–	Legendre Memory Units: Continuous-Time Representation in Recurrent Neural Networks	abr/neurips2019 · hrshtv/pytorch-lmu	2019
15	Adaptive-saturated RNN	96.96%	99.3%	Adaptive-saturated RNN: Remember more with less instability	ndminhkhoi46/asRNN	2023
16	Sparse Combo Net	96.94	–	RNNs of RNNs: Recursive Construction of Stable Assemblies of Recurrent Neural Networks	ennisthemennis/sparse-combo-net	2021
17	GAM-RHN-1	96.8%	–	Recurrent Highway Networks with Grouped Auxiliary Memory	WilliamRo/gam_rhn · MindCode-4/code-8 · MindSpore-scientific/code-5 · +1	2019
18	LEM	96.6%	99.5%	Long Expressive Memory for Sequence Modeling	tk-rusch/lem	2021
19	LipschitzRNN	96.3%	99.4	Lipschitz Recurrent Neural Networks	erichson/LipschitzRNN	2020
19	DNC+CUW	96.3%	99.1%	Learning to Remember More with Less Memorization	thaihungle/UW-DNC	2019