A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is
In the experiment, we empirically evaluate the generalization property using HSIC in both the reconstruction and prediction auto-encoding (AE) architectures. ... Recent works investigated the generalization properties in deep neural networks (DNNs) by studying the Information Bottleneck in DNNs. ... In the experiment, we empirically evaluate the generalization property of the learned latent representations in reconstruction and prediction auto-encoding (AE) architectures. ...arXiv:1802.05408v1 fatcat:sltamp7gdba3ngmvu2n5bvpxzm
, and Auto-encoder (AE)  . ... HSIC Bottleneck has been developed based on Hilbert-Schmidt Independence Criterion (HSIC)  , Information Bottleneck  , and Dependency Bottleneck  theorems. ...doi:10.1109/access.2020.3040083 fatcat:tsqokovkm5gpfdsnm7bph73piu
We present a meta-analysis of the performance of speech recognition adaptation algorithms, based on relative error rate reductions as reported in the literature. ...  used accent-dependent bottleneck features in place of i-vectors; and Turan et al.  used x-vector accent embeddings in a semi-supervised setting. ... The AED architecture ( Fig. 1(c) ) enriches the encoder-decoder model with an additional attention network which interfaces the acoustic encoder with the decoder. ...doi:10.1109/ojsp.2020.3045349 fatcat:tdprznnisvc2xampkzbavjpamu
Computer Communications and Networks
This chapter focuses on offering an overview of failures observed in real large-scale systems and their characteristics, with an emphasis on modeling, detection, and prediction. ... Understanding the behavior of failures in large-scale systems is important in order to design techniques to tolerate them. ... Firstly, it can highlight dependability bottlenecks and might serve as a guideline for designing more reliable systems in the future. ...doi:10.1007/978-3-319-20943-2_2 fatcat:44yvw6pozjbxlmhbz7kafvcapq