關鍵字查詢 | 類別:會議論文 | | 關鍵字:A knowledge-based approach to supervised incremental learning

[第一頁][上頁]1[次頁][最末頁]目前在第 1 頁 / 共有 01 筆查詢結果
序號 學年期 教師動態
1 82/2 資工系 許輝煌 教授 會議論文 發佈 A knowledge-based approach to supervised incremental learning , [82-2] :A knowledge-based approach to supervised incremental learning會議論文A knowledge-based approach to supervised incremental learningFu, Li-min; Hsu, Hui-huang; Principe, Jose C.淡江大學資訊工程學系Institute of Electrical and Electronics Engineers (IEEE)Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on, vol.3, pp.1793-1798How to learn new knowledge without forgetting old knowledge is a key issue in designing an incremental-learning neural network. In this paper, we present a rule-based connectionist approach in which old knowledge is preserved by bounding weight modifications. In addition, some heuristics are developed for avoiding overtraining of the network and adding new hidden units. The feasibility of this approach is demonstrated for classification problems including the iris and the promoter domains.tku_id:000117151;Made available in DSpace on 2010-04-15T01:45:51Z (GMT). No. of bitstreams: 1
 078031901X_3p1793-1798.pdf: 40
[第一頁][上頁]1[次頁][最末頁]目前在第 1 頁 / 共有 01 筆查詢結果