Autoencoder
释义 Definition
自编码器:一种神经网络模型,目标是把输入数据先“压缩”为较小的表示(编码),再尽量“还原”回原始输入(解码),常用于降维、特征学习、去噪与异常检测。(也有多种变体,如去噪自编码器、稀疏自编码器、变分自编码器等。)
例句 Examples
An autoencoder can compress images into a small vector.
自编码器可以把图像压缩成一个较小的向量表示。
By training an autoencoder on normal sensor readings, the system can flag anomalies when reconstruction error becomes unusually high.
将自编码器在正常传感器读数上训练后,当重建误差异常升高时,系统就能标记出异常情况。
发音 Pronunciation (IPA)
/ˌɔːtoʊɪnˈkoʊdər/
词源 Etymology
auto- 表示“自我、自动”,encoder 表示“编码器”。合起来字面意思是“自我编码/自我还原的编码器”,强调模型用自身学到的表示来重建输入数据。
相关词 Related Words
文学与著作 Literary & Notable Works
- Ian Goodfellow, Yoshua Bengio, Aaron Courville:《Deep Learning》
- Christopher M. Bishop:《Pattern Recognition and Machine Learning》
- Pascal Vincent et al. (2010): “Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion”
- Diederik P. Kingma & Max Welling (2013): “Auto-Encoding Variational Bayes”
- Geoffrey E. Hinton & Ruslan R. Salakhutdinov (2006): “Reducing the Dimensionality of Data with Neural Networks”