Add resource "The WeisfeilerLehman Isomorphism Test" Accepted
Changes: 9

Add The WeisfeilerLehman Isomorphism Test
 Title

 Unchanged
 The WeisfeilerLehman Isomorphism Test
 Type

 Unchanged
 Blog post
 Created

 Unchanged
 no value
 Description

 Unchanged
 Two graphs are considered isomorphic if there is a mapping between the nodes of the graphs that preserves node adjacencies. Here is the algorithm for the WeisfeilerLehman Isomorphism Test. It produces for each graph a canonical form. If the canonical forms of two graphs are not equivalent, then the graphs are definitively not isomorphic. However, it is possible for two nonisomorphic graphs to share a canonical form, so this test alone cannot provide conclusive evidence that two graphs are isomorphic.
 Link

 Unchanged
 https://davidbieber.com/post/20190510weisfeilerlehmanisomorphismtest/
 Identifier

 Unchanged
 no value
Resource  v1  current (v1) 
Add How Powerful are Graph Neural Networks?
 Title

 Unchanged
 How Powerful are Graph Neural Networks?
 Type

 Unchanged
 Paper
 Created

 Unchanged
 20190222
 Description

 Unchanged
 Despite GNNs revolutionizing graph representation learning, there is limited understanding of their representational properties and limitations. Here, we present a theoretical framework for analyzing the expressive power of GNNs to capture different graph structures. Our results characterize the discriminative power of popular GNN variants, such as Graph Convolutional Networks and GraphSAGE, and show that they cannot learn to distinguish certain simple graph structures. We then develop a simple architecture that is provably the most expressive among the class of GNNs and is as powerful as the WeisfeilerLehman graph isomorphism test.
 Link

 Unchanged
 http://arxiv.org/abs/1810.00826
 Identifier

 Unchanged
 no value
Resource  v1  current (v1) 
Add Reversals in psychology
 Title

 Unchanged
 Reversals in psychology
 Type

 Unchanged
 Blog post
 Created

 Unchanged
 20200126
 Description

 Unchanged
 Psychology has in recent years been racking up reversals: in fact only 4065% of its classic social results were replicated, in the weakest sense of finding ‘significant’ results in the same direction. (Even in those that replicated, the average effect found was half the originally reported effect.) The following are empirical findings about empirical findings; they’re all open to rereversal. Also it’s not that “we know these claims are false”: failed replications (or proofs of fraud) usually just challenge the evidence for a hypothesis, rather than affirm the opposite hypothesis.
 Link

 Unchanged
 https://www.gleech.org/psych
 Identifier

 Unchanged
 no value
Resource  v1  current (v1) 
Add Graph isomorphism problem
 Title

 Unchanged
 Graph isomorphism problem
 Description

 Unchanged
 The graph isomorphism problem is the computational problem of determining whether two finite graphs are isomorphic. The problem is not known to be solvable in polynomial time nor to be NPcomplete, and therefore may be in the computational complexity class NPintermediate. It is known that the graph isomorphism problem is in the low hierarchy of class NP, which implies that it is not NPcomplete unless the polynomial time hierarchy collapses to its second level. At the same time, isomorphism for many special classes of graphs can be solved in polynomial time, and in practice graph isomorphism can often be solved efficiently.
 Link

 Unchanged
 https://en.wikipedia.org/?curid=1950766
Topic  v1  current (v1) 
Add Graph convolutional networks (GCN) compared in How Powerful are Graph Neural Networks?
 Current
 compared in
Topic to resource relation  v1 
Add Psychology cons given in Reversals in psychology
 Current
 cons given in
Topic to resource relation  v1 
Add Graph convolutional networks (GCN) (detached) The WeisfeilerLehman Isomorphism Test
 Current
 (detached)
Topic to resource relation  v1 
Add Graph isomorphism problem treated in The WeisfeilerLehman Isomorphism Test
 Current
 treated in
Topic to resource relation  v1 
Add Graph convolutional networks (GCN) relates to Graph isomorphism problem
 Current
 relates to
Topic to topic relation  v1