神經網絡

  呢篇文章講嘅係神經網絡－無論生物定人工。想搵有關人工神經網絡嘅嘢，請睇「人工神經網絡」。

神經網絡（粵拼：san4 ging1 mong5 lok6；英文：neural network）係指由一柞神經細胞（neuron）組成嘅網絡，例如一個生物神經網絡由一柞生勾勾嘅神經細胞組成，每粒神經細胞會用電同化學物向第啲細胞傳訊號，而一個人工神經網絡（ANN）通常會用電腦程式等嘅方法模擬呢個過程，並且嘗試用嚟解一啲生物神經網絡解到、傳統電腦程式解唔到嘅問題^[1]。舉個例說明，例如家吓搵個（簡單嘅前饋型）神經網絡，搵其中一粒人工神經細胞集中睇佢，佢會有返個數字嚟反映佢嘅啟動程度，而呢個數字取決於第啲人工神經細胞嘅啟動程度－即係話個程式會有一條類似噉樣嘅算式：

${\displaystyle t=W_{1}A_{1}+W_{2}A_{2}...}$；

一個簡單、單向嘅神經網絡圖例

喺呢條式當中，${\displaystyle t}$ 代表嗰粒神經細胞嘅啟動程度，${\displaystyle A_{n}}$ 代表其他神經細胞當中第 ${\displaystyle n}$ 粒嘅啟動程度，而 ${\displaystyle W_{n}}$ 就係其他神經細胞當中第 ${\displaystyle n}$ 粒嘅權重（指嗰粒神經細胞有幾影響到 ${\displaystyle t}$ 嗰粒神經細胞嘅啟動程度）。所以當一粒人工神經細胞啟動嗰陣，會帶起佢後面啲人工神經細胞跟住佢啟動－似十足生物神經網絡入面嗰啲神經細胞噉。假如個神經網絡嘅程式有演算法令佢能夠自行按照經驗改變 ${\displaystyle W_{n}}$ 嘅數值嘅話，佢就會曉學習（根據經驗改變自己嘅行為）^[2][3][4]。

到咗廿一世紀頭，人工神經網絡經已俾科學界廣泛噉應用落去各種各樣嘅問題嗰度，例如係電腦視覺同埋語音識別呢啲傳統嘅電腦程式好難解決到嘅問題，都有得用人工神經網絡嘅方法應付^[5][6][7]。

睇埋

• 人工神經網絡，有陣時啲人會直接嗌人工神經網絡做「神經網絡」。
• 認知科學
• 認知模型
• 神經細胞
• 腦部
• 網絡理論

攷

1. Bryson, Arthur Earl (1969). Applied Optimal Control: Optimization, Estimation and Control. Blaisdell Publishing Company or Xerox College Publishing. p. 481.
2. Omidvar, O., & Elliott, D. L. (1997). Neural systems for control. Elsevier.
3. Tahmasebi; Hezarkhani (2012). "A hybrid neural networks-fuzzy logic-genetic algorithm for grade estimation". Computers & Geosciences. 42: 18–27
4. Hopfield, J. J. (1982). "Neural networks and physical systems with emergent collective computational abilities". Proc. Natl. Acad. Sci. U.S.A. 79 (8): 2554–2558.
5. Ivakhnenko, A. G.; Grigorʹevich Lapa, Valentin (1967). Cybernetics and forecasting techniques. American Elsevier Pub. Co.
6. Akira Hirose, Shotaro Yoshida (2012). “Generalization Characteristics of Complex-valued Feedforward Neural Networks in Relation to Signal Coherence”. IEEE TNNLS, 23(4): 541-551.
7. Forrest MD (April 2015). "Simulation of alcohol action upon a detailed Purkinje neuron model and a simpler surrogate model that runs >400 times faster". BMC Neuroscience. 16 (27): 27.

拎

• A Brief Introduction to Neural Networks (D. Kriesel) - Illustrated, bilingual manuscript about artificial neural networks; Topics so far: Perceptrons, Backpropagation, Radial Basis Functions, Recurrent Neural Networks, Self Organizing Maps, Hopfield Networks.
• Review of Neural Networks in Materials Science
• Artificial Neural Networks Tutorial in three languages (Univ. Politécnica de Madrid).
• Another introduction to ANN.
• Next Generation of Neural Networks - Google Tech Talks.
• Performance of Neural Networks.
• Neural Networks and Information.