brazofuerte
  • Member for 2 years, 8 months
  • Last seen this week
  • UK
5 answers
19 votes
8k views
What is the difference between latent and embedding spaces?
Accepted answer
12 votes

Embedding vs Latent Space Due to Machine Learning's recent and rapid renaissance, and the fact that it draws from many distinct areas of mathematics, statistics, and computer science, it often has a ...

View answer
4 answers
16 votes
10k views
What is non-Euclidean data?
Accepted answer
8 votes

I presume this question was prompted by the paper Geometric deep learning: going beyond Euclidean data (2017). If we look at its abstract: Many scientific fields study data with an underlying ...

View answer
1 answers
5 votes
848 views
What is a graph neural network?
3 votes

Graph Neural Networks The term Graph Neural Network, in its broadest sense, refers to any Neural Network designed to take graph structured data as its input: To cover a broader range of methods, this ...

View answer
2 answers
4 votes
5k views
What is feature embedding in the context of convolutional neural networks?
3 votes

The term feature embedding appears to be a synonym for feature extraction, feature learning etc. I.e. a form of embedding/dimension reduction (with the caveat the goal may not be a lower dimensional ...

View answer
1 answers
5 votes
97 views
Why are "Transformers" called this way?
2 votes

The authors of the original paper don't provide an explanation, but I suspect it's a combination of: popular recognizable branding (cf. BERT, DALL-E, Watson etc) similarity to [sequence] transduction ...

View answer
2 answers
5 votes
86 views
What are examples of approaches to dimensionality reduction of feature vectors?
2 votes

Some examples of dimensionality reduction techniques: Linear methods Non-linear methods Graph-based methods("Network embedding") PCA CCA ICA SVD LDA NMF Kernel PCA GDA Autoencoders t-SNE ...

View answer
2 answers
4 votes
452 views
Are there neural networks that accept graphs or trees as inputs?
2 votes

Yes, there are numerous, coming under the umbrella term Graph Neural Networks (GNN). The most common input structures accepted by these techniques are the adjacency matrix of the graph (optionally ...

View answer
2 answers
2 votes
197 views
Is ReLU a non-linear activation function?
2 votes

ReLU is non-linear by definition In calculus and related areas, a linear function is a function whose graph is a straight line, that is a polynomial function of degree one or zero. Since the graph ...

View answer
2 answers
8 votes
1k views
In what ways is the term "topology" applied to Artificial Intelligence?
2 votes

In addition to the ways the term topology is itself used generically to describe the "shape" of various aspects of Machine Learning, the term appears in the field Topological Data Analysis: In ...

View answer
2 answers
6 votes
495 views
Is there an open-source implementation for graph convolution networks for weighted graphs?
Accepted answer
2 votes

A Comprehensive Survey on Graph Neural Networks (2019) presents a list of ConvGNN's. All of the following accept weighted graphs, and three accept those with edge weights as well: And below is a ...

View answer
5 answers
36 votes
66k views
What is the difference between a convolutional neural network and a regular neural network?
2 votes

The everyday definition of convolution comes from the Latin convolutus meaning 'to roll together'. Hence the meaning twisted or complicated. The mathematical definition comes from the same root, with ...

View answer
2 answers
5 votes
396 views
What is different in each head of a multi-head attention mechanism?
1 votes

Multiple attention heads in a single layer in a transformer is analogous to multiple kernels in a single layer in a CNN: they have the same architecture, and operate on the same feature-space, but ...

View answer
3 answers
10 votes
887 views
What do you call a machine learning system that keeps on learning?
1 votes

ML, being a relatively young and fast-developing field, has numerous (near-)synonyms for many concepts. One paradigm difference is whether a model is learned from a static, pre-defined set of data, or ...

View answer
3 answers
1 votes
118 views
Are there any public real-life code examples of ML applications in Python?
1 votes

If you search Papers with Code for python "machine learning" (or a more specific query) you will get numerous results. Note these will be mostly scientific applications or methods.

View answer
1 answers
2 votes
56 views
What is the difference between graph semi-supervised learning and normal semi-supervised learning?
1 votes

The authors of your cited paper use the term graph-based semi-supervised learning (G-SSL) to refer to semi-supervised learning techniques which take graph structured data as their input. Given their ...

View answer
3 answers
5 votes
132 views
Is there a Python API for manipulating publicly available datasets?
1 votes

There are multiple python packages with inbuilt toy-datasets for testing purposes: sklearn.datasets seaborn.load_dataset() statsmodels.api.datasets rpy2 (requires R and pandas) PyDataset

View answer
1 answers
4 votes
114 views
What are the exact meaning of "lower-order structure" and "higher-order structure" in this paper?
1 votes

Low order/low level information refers to the most granular level of information. This is the most informative in terms of volume of information, but it can often be difficult to conceptualise for ...

View answer
4 answers
11 votes
3k views
What are the purposes of autoencoders?
1 votes

The decoder half is necessary in order to compute the loss function for training the network. Similar to how the 'adversary' is still necessary in a GAN even if you are only interested in the ...

View answer
1 answers
7 votes
175 views
A* is similar to Dijkstra with reduced cost
1 votes

What you are doing when calculating $d'(x,y)$: $d(x,y)$: calculating the original edge distance from $x$ to $y$ $h(y)$: plus the heuristic from $y$ to the goal $h(x)$: minus the heuristic from $x$ to ...

View answer
2 answers
3 votes
263 views
How to represent and work with the feature matrix for graph convolutional network (GCN) if the number of features for each node is different?
0 votes

My immediate suggestion would be to zero-fill the missing values, but I recalled the below comment suggesting a more sophisticated method: Karim: How to deal with different size of feature vectors? ...

View answer
2 answers
4 votes
122 views
What is the proof that the branch and bound algorithm always finds optimal path in a graph?
0 votes

Branch and Bound is similar to an exhaustive search, except it incorporates a method for computing lower bounds on branches. If the lower bound on a given branch is greater than the upper bound on the ...

View answer