2022
DOI: 10.48550/arxiv.2203.13534
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Generalization bounds for learning under graph-dependence: A survey

Abstract: Traditional statistical learning theory relies on the assumption that data are identically and independently generated from a given distribution (i.i.d.). The independently distributed assumption, on the other hand, fails to hold in many real applications. In this survey, we consider learning settings in which examples are dependent and their dependence relationship can be characterised by a graph. We collect various graph-dependent concentration bounds, which are then used to derive Rademacher and stability g… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
2
0

Year Published

2022
2022
2023
2023

Publication Types

Select...
1
1

Relationship

0
2

Authors

Journals

citations
Cited by 2 publications
(2 citation statements)
references
References 20 publications
0
2
0
Order By: Relevance
“…A tree-partition is a T-partition for some tree T. The tree-partition-width 3 of G, denoted by tpw(G), is the minimum width of a tree-partition of G. Thus tpw(G) = tpw 1 (G), which equals the minimum ∈ N 0 for which G is contained in T K for some tree T. Tree-partitions were independently introduced by Seese [69] and Halin [39], and have since been widely investigated [7,8,19,20,24,34,76,77]. Applications of tree-partitions include graph drawing [13,16,30,32,80], nonrepetitive graph colouring [2], clustered graph colouring [1], monadic second-order logic [51], network emulations [4,5,9,37], size Ramsey number [26,43], statistical learning theory [81], and the edge-Erdős-Pósa property [14,38,60]. Planar-partitions and other more general structures have also been studied [18,21,22,63,80].…”
Section: Partitionsmentioning
confidence: 99%
“…A tree-partition is a T-partition for some tree T. The tree-partition-width 3 of G, denoted by tpw(G), is the minimum width of a tree-partition of G. Thus tpw(G) = tpw 1 (G), which equals the minimum ∈ N 0 for which G is contained in T K for some tree T. Tree-partitions were independently introduced by Seese [69] and Halin [39], and have since been widely investigated [7,8,19,20,24,34,76,77]. Applications of tree-partitions include graph drawing [13,16,30,32,80], nonrepetitive graph colouring [2], clustered graph colouring [1], monadic second-order logic [51], network emulations [4,5,9,37], size Ramsey number [26,43], statistical learning theory [81], and the edge-Erdős-Pósa property [14,38,60]. Planar-partitions and other more general structures have also been studied [18,21,22,63,80].…”
Section: Partitionsmentioning
confidence: 99%
“…Tree-partitions were independently introduced by Seese [68] and Halin [38], and have since been widely investigated [7,8,19,20,31,75,76]. Applications of tree-partitions include graph drawing [12,15,28,30,79], nonrepetitive graph colouring [2], clustered graph colouring [1], monadic second-order logic [50], network emulations [4,5,9,36], statistical learning theory [80], and the edge-Erdős-Pósa property [13,37,59]. Planar-partitions and other more general structures have also been studied [17,21,22,62,79].…”
Section: Partitionsmentioning
confidence: 99%