Random features for Kernel Deep Convex Network

By Po-Sen Huang, Li Deng, Mark Hasegawa-Johnson, Xiaodong He. 2013

42 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Dong Yu
2. Li Deng
3. Yifan Gong
4. Jinyu Li
5. Jianfeng Gao
6. Po-Sen Huang
7. Hamid Palangi
8. Yoshua Bengio

In Signal Processing

1. Dong Yu
2. Li Deng
3. Yifan Gong
4. Jinyu Li
5. Deliang Wang
6. Paris Smaragdis
7. Po-Sen Huang
8. Rabab Kreidieh Ward

Citations in the corpus

• Deep Learning: Methods and Applications (2014, methodology): “[171], based on the use of random Fourier features, which possess the strong theoretical property of approximating the Gaussian kernel while rendering efficient computation in both training and evaluation of the K-DSN with large training samples.”
• Predicting speech recognition confidence using deep learning with word identity and score features (2013, methodology): “Kernel Deep Convex Network (K-DCN) was proposed in [10, 11].”
• A Review of Deep Learning Models for Time Series Prediction (2019, methodology): “Motivated by increase the size of the hidden units without increasing the number of parameters to learn, Kernel DSN (KDSN) has been proposed using kernel trick [125].”
• Ensemble deep learning: A review (2021, methodology): “To leverage the memory requirements, random Fourier feature-based kernel deep convex network [82] approximated the Gaussian kernel which reduces the training time and helps in”
• Using deep stacking network to improve structured compressed sensing with Multiple Measurement Vectors (2013, background): “Other applications of DSNs are introduced in [18] and [19].”
• What are Extreme Learning Machines? Filling the Gap Between Frank Rosenblatt’s Dream and John von Neumann’s Puzzle (2015, methodology): “Similarly, there is nothing wrong to have new terms for the variants of ELM (with Fourier series nodes) with significant extensions (such as Random Kitchen Sink (RKS) [10], RKS’ further extension - FastFood [11] and Convex Network [12]) as well as ELM with LMS referred to as No-Prop algorithm [13].”
• Deep learning for monaural speech separation (2014, background): “RELATION TO PREVIOUS WORK Deep learning approaches have yielded many state of the art results by representing different levels of abstraction with multiple nonlinear layers [8, 11, 12].”
• Kernel methods match Deep Neural Networks on TIMIT (2014, unknown): “proposed a kernel deep convex network [10, 11].”