本頁只刊出中文翻譯與中文說明；英文原文請見下方原文連結。
原文連結
論文資訊
- 類型：已發表論文
- 日期：2020
摘要
The ability to store and manipulate 資訊 is a hallmark of 計算 systems. Whereas computers are carefully engineered to represent and perform 數學 operations on struc- tured data, neuro生物 systems perform analogous functions despite flexible organization and unstructured sensory input. Recent efforts have made progress in modeling the representation and recall of 資訊 in 神經 systems. However, precisely how 神經 systems learn to modify these representations remains far from understood. Here we demonstrate that a recurrent 神經 網絡 (RNN) can learn to modify its representation of complex 資訊 using only exam- ples, and we explain the associated learning mechanism with new theory. Specifically, we drive an RNN with examples of translated, linearly transformed, or pre-bifurcated time series from a chaotic Lorenz
※ 此為已發表論文，全文需透過期刊付費取得