知识库｜湖南麓川信息科技有限公司

--- deep_learning:incremental_learning:incremental_learning_papers [2023/03/08 15:56]
xujianglong ↷ Page moved from deep_learning:incremental_learning:incremental_learning_papers to 内部资料:deep_learning:incremental_learning:incremental_learning_papers
+++ deep_learning:incremental_learning:incremental_learning_papers [2023/03/08 16:11] (current)
xujianglong ↷ Links adapted because of a move operation
@@ Line 261: / Line 261: @@
 **Joint Training**: **All parameters $\theta_s$, $\theta_o$, $\theta_n$ are jointly optimized.**  **Drawback**: Joint training becomes increasingly cumbersome in training as more tasks are learned and is not possible if the training data for previously learned tasks is unavailable.
-{{:deep_learning:incremental_learning:illustration_lwf_others.png}}
+{{deep_learning:incremental_learning:illustration_lwf_others.png}}
 **Algorithm of LwF**
-{{:deep_learning:incremental_learning:procedure_lwf.png}}
+{{deep_learning:incremental_learning:procedure_lwf.png}}
 **Backbone**
@@ Line 303: / Line 303: @@
 Classification
-{{:deep_learning:incremental_learning:icarl_algorithm1.png?nolink&405x242}}
+{{deep_learning:incremental_learning:icarl_algorithm1.png?nolink&405x242}}
 Training
-{{:deep_learning:incremental_learning:icarl_algorithm2.png?nolink&398x316}}
+{{deep_learning:incremental_learning:icarl_algorithm2.png?nolink&398x316}}
 **Why nearest-mean-of-exemplars classification**