0
2
1748
1
評價神經(jīng)網(wǎng)絡(luò): 現(xiàn)有問題:學(xué)得慢、學(xué)得不是真正規(guī)律(有干擾) 訓(xùn)練數(shù)據(jù)70%吕喘;測試數(shù)據(jù)30% 誤差曲線砸抛,(分類)精確度曲線+(回歸問題)R2分?jǐn)?shù)...
"?two branches for Deep Reinforcement Learning: based on Value or Policy...