| 3 | 1/1 | 返回列表 |
| 查看: 705 | 回復(fù): 2 | |||
robbie82新蟲 (初入文壇)
|
[交流]
【求助】如何回答審稿人的意見? 已有2人參與
|
|
剛收到IJAMT的審稿人意見,建議major revision,下面是審稿人意見原文,有些不明白的地方,向蟲友們請(qǐng)教,我的paper中用了神經(jīng)網(wǎng)絡(luò)的BP算法 1 How do you minimized the overfitting ( since it requires a great deal of computational effort)? 2 How do you overcome the black box of the model since the individual relations between the input variables and the output variables are not developed by engineering judgment? 3 Back propagation is famous for the slow rate of convergence, which requires very long training times, what is the your course of action in this study? minimize overfitting一般從哪些方面入手呢?可參考哪些文獻(xiàn)呢? 從哪些方面談如何overcome問題2中的black box比較好呢?可參考哪些文獻(xiàn)呢? 問題3是不是表明審稿人認(rèn)為用BP算法解決我提出的問題不合適? |
|
這個(gè)審稿人好對(duì)付啊。 訓(xùn)練時(shí)間最好說,就說網(wǎng)絡(luò)結(jié)構(gòu)不大,訓(xùn)練時(shí)間雖然長(zhǎng),但還能接受。 overfitting也好辦,隨便找個(gè)辦法都行。比如把訓(xùn)練集分出一塊來,不參加訓(xùn)練,當(dāng)它的準(zhǔn)確率開始下降時(shí),停止訓(xùn)練。 blackbox嘛你就說訓(xùn)練完了就用,現(xiàn)在的工作不分析這個(gè)input output 的relationship。他要是不依不饒,以后還有第二回合,隨便找個(gè)分析工具送給他就行了。 |
銀蟲 (小有名氣)
|
1 How do you minimized the overfitting ( since it requires a great deal of computational effort)? 可以采用early stop或者k-fold cross validation。估計(jì)你的文章中沒有對(duì)過擬合討論,所以就算性能再好,也沒有意義。 2 How do you overcome the black box of the model since the individual relations between the input variables and the output variables are not developed by engineering judgment? 這個(gè)主要就是詳細(xì)著重闡述,如果采用專家系統(tǒng)、決策樹來推導(dǎo)輸入輸出關(guān)系,那么擬合的性能就會(huì)下降。為了提高性能,采用神經(jīng)網(wǎng)絡(luò),雖然導(dǎo)致輸入輸出關(guān)系模糊,但是是值得的。 3 Back propagation is famous for the slow rate of convergence, which requires very long training times, what is the your course of action in this study? 這個(gè)就麻煩了,因?yàn)锽P算法確實(shí)性能很差,建議你改用其他算法重做一下。 |
| 3 | 1/1 | 返回列表 |
| 最具人氣熱帖推薦 [查看全部] | 作者 | 回/看 | 最后發(fā)表 | |
|---|---|---|---|---|
|
[考研] 317求調(diào)劑 +7 | 十閑wx 2026-03-24 | 7/350 |
|
|---|---|---|---|---|
|
[考研] 303求調(diào)劑 +5 | DLkz1314. 2026-03-30 | 5/250 |
|
|
[考研] 一志愿哈爾濱工業(yè)大學(xué)材料與化工方向336分 +12 | 辰沐5211314 2026-03-26 | 12/600 |
|
|
[考研]
|
gr哈哈哈 2026-03-28 | 3/150 |
|
|
[考研] 085600 295分求調(diào)劑 +5 | W55j 2026-03-30 | 7/350 |
|
|
[考研] 材料專碩調(diào)劑 +11 | 椰椰。 2026-03-29 | 11/550 |
|
|
[考研] 303求調(diào)劑 +7 | DLkz1314. 2026-03-30 | 7/350 |
|
|
[考研] 求化學(xué)調(diào)劑 +11 | wulanna 2026-03-28 | 11/550 |
|
|
[考研] 求調(diào)劑,一志愿 南京航空航天大學(xué)大學(xué) ,080500材料科學(xué)與工程學(xué)碩 +6 | @taotao 2026-03-26 | 7/350 |
|
|
[考研] 085600,材料與化工321分求調(diào)劑 +10 | 大饞小子 2026-03-28 | 10/500 |
|
|
[考研] 316求調(diào)劑 +7 | 江辭666 2026-03-26 | 7/350 |
|
|
[考研] 322求調(diào)劑 +5 | 舊吢 2026-03-24 | 5/250 |
|
|
[考研] 085701環(huán)境工程,267求調(diào)劑 +16 | minht 2026-03-26 | 16/800 |
|
|
[考研] 調(diào)劑 +3 | 好好讀書。 2026-03-28 | 3/150 |
|
|
[考研] 085602 307分 求調(diào)劑 +7 | 不知道叫什么! 2026-03-26 | 7/350 |
|
|
[考研] 265求調(diào)劑 +8 | 小木蟲085600 2026-03-27 | 8/400 |
|
|
[考研] 279 分 求調(diào)劑 +4 | 睡個(gè)好覺_16 2026-03-24 | 4/200 |
|
|
[考研] 081200-11408-276學(xué)碩求調(diào)劑 +3 | 崔wj 2026-03-26 | 3/150 |
|
|
[考研] 打過很多競(jìng)賽,085406控制工程300分,求調(diào)劑 +3 | askeladz 2026-03-26 | 3/150 |
|
|
[考研] 0854人工智能方向招收調(diào)劑 +4 | 章小魚567 2026-03-24 | 4/200 |
|