Carview!

CARVIEW

MOTORHOMES

Select Language

HTTP/2 200 server: GitHub.com content-type: application/xml last-modified: Thu, 26 May 2022 10:55:21 GMT access-control-allow-origin: * etag: W/"628f5c99-2bd2" expires: Mon, 29 Dec 2025 11:05:51 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: 3CDF:328FD3:8B77D8:9C8CDF:69525E36 accept-ranges: bytes age: 0 date: Mon, 29 Dec 2025 10:55:51 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210074-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1767005751.094789,VS0,VE228 vary: Accept-Encoding x-fastly-request-id: 02e5e195cd329b8bf03c41d2d5e193b3f4bd074b content-length: 3476 Jekyll2022-05-26T03:55:15-07:00https://panford.github.io/feed.xmlKobby Panford-Quainoopersonal descriptionKobby Panford-QuainooGetting Started with Graph Neural Networks2021-05-30T00:00:00-07:002021-05-30T00:00:00-07:00https://panford.github.io/posts/2021/05/blog-post-1Here, you’ll find a few resources that’ll get you started with Graph Neural Networks. I will keep updating it so you can always check back!. Intro to graph representation learning <ol> <li>Representation learning on graphs: Methods and Applications; <a href="https://arxiv.org/abs/1709.05584">Hamilton et al.</a></li> <li>Graph Neural Networks: A review or methods and applications; <a href="https://arxiv.org/abs/1812.08434">Zhou et al;</a></li> <li>Deep Learning on graphs: A survey; <a href="https://arxiv.org/pdf/1812.04202">Ziwei Zhang et al</a></li> </ol> Books: <ol> <li>Graph Representation learning book; <a href="https://www.cs.mcgill.ca/~wlh/grl_book/">Will Hamilton (Highly recommended for comprehensive understanding)</a></li> <li>Geometric Deep Learning Grids, Groups, Graphs, Geodesics, and Gauges; <a href="https://arxiv.org/pdf/2104.13478.pdf">Bronstein et al</a></li> </ol> Video: <ol> <li>Machine Learning with Graphs; Jure Leskovec (Stanford University), <a href="https://www.youtube.com/playlist?list=PLoROMvodv4rPLKxIpqhjhPgdQy7imNkDn">Videos</a>, <a href="https://web.stanford.edu/class/cs224w/">Course website</a></li> </ol> Important to know architectures <ol> <li>Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering (ChebNet); <a href="https://arxiv.org/pdf/1606.09375">Deferrard et al</a></li> <li>Graph Convolutional Neural Networks; <a href="https://arxiv.org/pdf/1609.02907">Kipf et al.</a></li> <li>Graph Attention Networks (GAT); <a href="https://arxiv.org/pdf/1710.10903">Velickovic et al.</a></li> <li>Inductive Representation learning in large graphs; <a href="https://arxiv.org/pdf/1706.02216">Hamilton et al</a></li> <li>Graph Isormorphism Networks; <a href="https://arxiv.org/pdf/1810.00826">Xu et al.</a></li> <li>Gated Graph Sequence Neural Networks; <a href="https://arxiv.org/abs/1511.05493">Li et al</a></li> <li>Benchmarking Graph Neural Networks; <a href="https://arxiv.org/pdf/2003.00982">Dwivedi et al</a></li> </ol> Frameworks/ tools <ol> <li>PyTorch Geometric <a href="https://pytorch-geometric.readthedocs.io/en/latest/notes/introduction.html">(Based on pytorch)</a></li> <li>Spekral <a href="https://graphneural.network/">(Tensorflow/ keras)</a></li> <li>Deep Graph Library <a href="https://www.dgl.ai/">(Framework agnostic, compatible with PyTorch, Tensorflow, MxNet etc)</a></li> <li>Open Graph Benchmark: <a href="https://arxiv.org/pdf/2005.00687">Datasets for Machine Learning on graphs</a></li> </ol>Kobby Panford-Quainoo

Here, you’ll find a few resources that’ll get you started with Graph Neural Networks. I will keep updating it so you can always check back!.

Adversarial Autoencoders2021-05-30T00:00:00-07:002021-05-30T00:00:00-07:00https://panford.github.io/posts/2013/08/blog-post-2<h2 id="short-notes-on-adversarial-autoencoders">Short Notes on Adversarial Autoencoders</h2> The Adversarial autoencoder (AAE) was proposed by <a href="https://arxiv.org/abs/1511.05644">Alireza Makhzani et al (2016)</a>. AAE is a generative model with two main objectives. First is to learn a compressed/ latent representation of data while ensuring that the learned latent representation follows a prior distribution we get to define. For instance, with an autoencoder, we have en encoder block that encodes the input data into a fixed vector $z$ of potentially lower dimension that the input and retaining as much information as possible. If we want to retrieve the input data, we simply pass the latent vector ,$z$, to a decoder which attempts to create the data (reconstruct) into the original dimension. One important point here is that the latent vector $z$ learned by the autoencoder is always fixed and the reconstructed data is pretty much the same each time the decoder reconstructs it. The $z$ is therefore useful if we want only to do dimensionality reduction. For variational autoencoders, instead of encoding the input data into a fixed vector ,$z$, the encoder learns parameters $\mu$ and $\sigma$ of a distribution which can be reparameterized with samples from a simpler distribution say a standard gaussian (a gaussian distribution with mean 0 and standard deviation 1) to get the z vector ($z = \mu + \sigma \epsilon $, where $\epsilon \sim \mathcal{N}(0,1)$). This allows us to create new forms of the data by simply sampling from the standard gaussian and have the decoder reconstruct the data. Sometimes however, the prior distribution might be different from the learned distribution. Reconstructing the data from samples from the prior might therefore not be very useful. The Adversarial autoencoder uses quite a bit different technique to impose the prior distribution on the latent code and besides, it allow some flexibility with the choice of prior distribution. By imposing a prior distribution on the latent code, we mean, we want the latent code to follow a prior distribution, say a gaussian distribution with mean 0 and standard deviation 1. <img src="/images/aae1.png" alt="Adversarial Autoencoder-complete architecture" /> Fig 1. Adversarial Autoencoder architecture  The AAE is made up of a generator and a discriminator as in the generative adversarial network. The generator is made up of an encoder and a decoder as for an autoencoder and the discriminator is a simple neural network which produces a binary output (0/1). Again, main objectives of the adversarial autoencoder is 1.) for the generator’s encoder to learn better representation for the input data and the decoder to be able to reconstruct the data as close as possible to the original input from the latent code. 2.) The discriminator should guide the generator’s encoder to produce a latent code that resembles samples from our prior distribution. Below, lets look at the two main training phases; <img src="/images/aae2.png" alt="Reconstruction Phase" /> Fig 2. Reconstruction Phase: update generator (encoder + decoder) Phase 1: The generator’s encoder encodes the input data into a latent representation whiles the decoder attempts to reconstruct the original data from the latent code. This represents the first training phase where the generator’s encoder and decoder are both optimized jointly to minimize reconstruction error in the generator. <img src="/images/aae3.png" alt="Regularization Phase: Discriminator training" /> Fig 3. Regularization phase: discriminator training Phase 2: i) The discriminator receives samples from the prior distribution and the latent code labeled as true/positive samples and false/ negative samples respectively. The samples are fed to the discriminator which is a neural network model with a binary output for this classification task. Only the discriminator’s parameters are updated at this point. <img src="/images/aae4.png" alt="Regularization Phase: Generator (Encoder) training" /> Fig 4. Regularization phase: encoder (generator) training ii) In the second part of the regularization phase, the generator’s encoder trains to fool the discriminator to believe that the samples it has generated come from the prior distribution. If the latent code truly resembles samples from the prior distribution, then the generator yields a low loss and a high loss if otherwise. Gradients from the discriminator is propagated to the generator’s encoder to update its parameters (implementation detail: discriminator’s parameters should be frozen at this point).  Kobby Panford-Quainoo

Short Notes on Adversarial Autoencoders

Original Source | Taken Source