Tasks
SotA
Datasets
Papers
Methods
Submit
About
SotA
/
Methodology
/
Incremental Learning
/
CIFAR-100 - 50 classes + 5 steps of 10 classes
Incremental Learning on CIFAR-100 - 50 classes + 5 steps of 10 classes
Metric: Average Incremental Accuracy (higher is better)
Leaderboard
Dataset
Loading chart...
Results
Submit a result
Export CSV
Sort:
Average Incremental Accuracy (best first)
Average Incremental Accuracy (worst first)
Date (newest first)
Date (oldest first)
Model name (A→Z)
#
Model
↕
Average Incremental Accuracy
▼
Augmentations
Paper
Date
↕
Code
1
TCIL
74.88
No
Resolving Task Confusion in Dynamic Expansion Ar...
2022-12-29
Code
2
TCIL-Lite
74.3
No
Resolving Task Confusion in Dynamic Expansion Ar...
2022-12-29
Code
3
DER(Standard ResNet-18)
72.6
No
DER: Dynamically Expandable Representation for C...
2021-03-31
Code
4
D3Former
72.23
No
D3Former: Debiased Dual Distilled Transformer fo...
2022-07-25
Code
5
FOSTER
69.46
No
FOSTER: Feature Boosting and Compression for Cla...
2022-04-10
Code
6
RMM (Modified ResNet-32)
68.86
No
RMM: Reinforced Memory Management for Class-Incr...
2023-01-14
Code
7
DER(Modified Res-32)
67.6
No
DER: Dynamically Expandable Representation for C...
2021-03-31
Code
8
CCIL-SD
67.17
No
Essentials for Class Incremental Learning
2021-02-18
Code
9
PODNet (CNN)
64.83
No
PODNet: Pooled Outputs Distillation for Small-Ta...
2020-04-28
Code
10
UCIR (CNN)*
63.42
No
-
-
Code
11
UCIR (NME)*
63.12
No
-
-
Code
12
iCaRL*
57.17
No
iCaRL: Incremental Classifier and Representation...
2016-11-23
Code
13
BiC
56.86
No
Large Scale Incremental Learning
2019-05-30
Code
#1
TCIL
SOTA
74.88
Average Incremental Accuracy
· 2022-12-29
Resolving Task Confusion in Dynamic Expansion Architectures for Class Incremental Learning
Code
#2
TCIL-Lite
74.3
Average Incremental Accuracy
· 2022-12-29
Resolving Task Confusion in Dynamic Expansion Architectures for Class Incremental Learning
Code
#3
DER(Standard ResNet-18)
SOTA
72.6
Average Incremental Accuracy
· 2021-03-31
DER: Dynamically Expandable Representation for Class Incremental Learning
Code
#4
D3Former
72.23
Average Incremental Accuracy
· 2022-07-25
D3Former: Debiased Dual Distilled Transformer for Incremental Learning
Code
#5
FOSTER
69.46
Average Incremental Accuracy
· 2022-04-10
FOSTER: Feature Boosting and Compression for Class-Incremental Learning
Code
#6
RMM (Modified ResNet-32)
68.86
Average Incremental Accuracy
· 2023-01-14
RMM: Reinforced Memory Management for Class-Incremental Learning
Code
#7
DER(Modified Res-32)
67.6
Average Incremental Accuracy
· 2021-03-31
DER: Dynamically Expandable Representation for Class Incremental Learning
Code
#8
CCIL-SD
SOTA
67.17
Average Incremental Accuracy
· 2021-02-18
Essentials for Class Incremental Learning
Code
#9
PODNet (CNN)
SOTA
64.83
Average Incremental Accuracy
· 2020-04-28
PODNet: Pooled Outputs Distillation for Small-Tasks Incremental Learning
Code
#10
UCIR (CNN)*
63.42
Average Incremental Accuracy
No paper
Code
#11
UCIR (NME)*
63.12
Average Incremental Accuracy
No paper
Code
#12
iCaRL*
SOTA
57.17
Average Incremental Accuracy
· 2016-11-23
iCaRL: Incremental Classifier and Representation Learning
Code
#13
BiC
56.86
Average Incremental Accuracy
· 2019-05-30
Large Scale Incremental Learning
Code