人工智慧 / DeepLearning
◎ 如果有一層隱藏層學不會的事,那就再加一層!
2016 年的春天,Google DeepMind 團隊的 AlphaGo 專案打敗了世界棋手,AI 世代的 Deep Learning 威力讓全世界 70 億人都驚呆了!
屬於機器學習一種的深度學習,透過層層類神經網路,儼然成為了最簡單卻又最深奧的技術,最早用於電腦視覺(Computer Vision)。現今主流還有應用於自然語言處理的RNN,可處理時間序列、判斷語意,以及近期最夯的生成式對抗網路(GAN)、及新興起的強化學習(Reinforcement Learning)。
◎ 那麼,如果有 N 層隱藏層學不會的事,那 N+1 層就會了嗎...
雖說深度學習強在他的「深」,但目前的研究結果也顯示了並非有越多層、成效就越好,從CNN中的梯度下降法(Gradient Descent Method)來看,當迭代過多層神經網路,容易導致Vanishing Gradient,在哪邊都看不出最小值。目前看來,要降低錯誤率還是需要較複雜的網路架構。
...說了這麼多,要進行深度學習,電腦需要的只是足夠量的資料,而你,需要的只是幾本好書。
相關書籍
-
95折
$3,230Deep Learning: Foundations and Concepts (Hardcover) -
$1,800$1,764 -
VIP 95折
$1,800$1,710 -
VIP 95折
$1,750$1,663 -
VIP 95折
$2,670$2,537 -
VIP 90折
$2,641$2,502 -
VIP 95折
$2,170$2,062 -
95折
$4,418Deep Reinforcement Learning for Wireless Communications and Networking: Theory, Applications and Implementation (Hardcover) -
90折
$1,935Designing Deep Learning Systems: A Software Engineer's Guide -
VIP 95折
$2,150$2,043 -
VIP 95折
$2,800$2,660 -
90折
$1,944Enhancing Deep Learning with Bayesian Inference: Create more powerful, robust deep learning systems with Bayesian deep learning in Python -
90折
$1,719Natural Language Understanding with Python: Combine natural language technology, deep learning, and large language models to create human-like language comprehension in computer systems (Paperback) -
VIP 90折
$2,508$2,376 -
VIP 95折
$2,200$2,090 -
VIP 95折
$2,250$2,138 -
VIP 95折
$1,710$1,625 -
90折
$1,620Hands-On Graph Neural Networks Using Python: Practical techniques and architectures for building powerful graph and deep learning apps with PyTorch (Paperback) -
VIP 90折
$2,565$2,430 -
$1,980$1,940 -
VIP 95折
$1,700$1,615 -
VIP 95折
$2,600$2,470 -
VIP 90折
$2,166$2,052 -
$1,780$1,744 -
VIP 90折
$1,245$1,179 -
$1,980$1,940 -
$1,980$1,940 -
VIP 95折
$2,150$2,043 -
85折
$3,757Zeroing Neural Networks -
VIP 95折
$2,070$1,967