Skip to Main Content
Table 4:
Current Task Accuracy for the Different Models for the SSMNIST Task Distribution.
Task Idsmall-Lstm-Gem-Net2Netlarge-Lstmlarge-Lstm-Gem
89.71 90.59 90.59 
77.67 86.71 86.33 
73.86 88.08 86.51 
74.437 88.12 84.68 
71.14 89.32 79.13 
67.84* 90.5  
61.24 90.97  
 90.3  
 89.89  
10  88.49  
11  81.64  
12  74.4  
13  67.8  
Task Idsmall-Lstm-Gem-Net2Netlarge-Lstmlarge-Lstm-Gem
89.71 90.59 90.59 
77.67 86.71 86.33 
73.86 88.08 86.51 
74.437 88.12 84.68 
71.14 89.32 79.13 
67.84* 90.5  
61.24 90.97  
 90.3  
 89.89  
10  88.49  
11  81.64  
12  74.4  
13  67.8  

Notes: The row with the asterisk denotes the task at which the proposed model expanded. Capacity expansion technique allows our proposed model to clear more tasks than it would have cleared otherwise. The proposed small-Lstm-Gem-Net2Net model clears more levels than the large-Lstm-Gem model.

Close Modal

or Create an Account

Close Modal
Close Modal