Novel applications of Machine Learning to Network Traffic Analysis

PDF Publication Title:

Novel applications of Machine Learning to Network Traffic Analysis ( novel-applications-machine-learning-network-traffic-analysis )

Previous Page View | Next Page View | Return to Search List

Text from PDF Page: 170

Figure 2. Details on the loss function elements for the VGM model. 3.3. Variants of the model We have explored three options for the general VGM model discussed in the previous section. 3.3.1. Option A. Model with Gaussian and Bernoulli distributions In this option we have the vector of labels as input to the network and the vector of features as output. Figure 3 presents option A. We use a multivariate Gaussian as the distribution for ๐’’(๐’/๐‘ณ), with a mean ๐(๐‘ณ)โกand a diagonal covariance matrix: ๐œฎ(๐‘ณ) โ†’ โกโก ๐ˆ๐Ÿ(๐‘ณ),โกwith different values along the diagonal. We ๐’Š have a standard normal ๐‘ต(๐ŸŽ, ๐‘ฐ) as the prior distribution for Z. ฬ‚ For the distribution ๐ฉ(๐—/๐™) we use a multivariate Bernoulli distribution. The nice property about the Bernoulli distribution is that we do not require doing sampling on it, as the output parameter that characterizes the distribution is the mean that is the same as the probability of ฬ‚ success. Then, in this case, the output of the last layer is taken as our final output ๐‘ฟ. ฬ‚ The selection of distributions for ๐’’(๐’/๐‘ณ) and ๐ฉ(๐—/๐™) is similar to the distributions selected in [1], since they are simple and provide good results. In Figure 3 we show also (in squared boxes) the elements of the loss function to be minimized for this option. Doctoral Thesis: Novel applications of Machine Learning to NTAP - 168

PDF Image | Novel applications of Machine Learning to Network Traffic Analysis

PDF Search Title:

Novel applications of Machine Learning to Network Traffic Analysis

Original File Name Searched:

456453_1175348.pdf

DIY PDF Search: Google It | Yahoo | Bing

Cruise Ship Reviews | Luxury Resort | Jet | Yacht | and Travel Tech More Info

Cruising Review Topics and Articles More Info

Software based on Filemaker for the travel industry More Info

The Burgenstock Resort: Reviews on CruisingReview website... More Info

Resort Reviews: World Class resorts... More Info

The Riffelalp Resort: Reviews on CruisingReview website... More Info

CONTACT TEL: 608-238-6001 Email: greg@cruisingreview.com (Standard Web Page)