RelationalGCNOptions<T>
Options & Configuration · AiDotNet.Models.Options
Configuration options for RelationalGCN (Relational Graph Convolutional Network).
Inheritance: ModelOptions → RegressionOptions<T> → TimeSeriesRegressionOptions<T> → RelationalGCNOptions<T>
For Beginners
RelationalGCN is designed for knowledge graphs and multi-relational data:
The Key Insight: Standard GCN treats all edges equally, but in financial networks, different types of relationships matter differently. A “supplier-of” relationship is different from a “competitor-of” relationship. R-GCN learns separate transformations for each relation type.
What Problems Does RelationalGCN Solve?
- Entity classification in knowledge graphs (company type, sector classification)
- Link prediction in multi-relational networks (predicting missing relationships)
- Financial network analysis with multiple relationship types
- Supply chain modeling with different connection types
How RelationalGCN Works:
- Relation-Specific Weights: Learns different weights for each relation type
- Basis Decomposition: Efficiently shares parameters across relations
- Block Decomposition: Alternative parameter sharing using block-diagonal matrices
- Self-Connections: Special weight for node’s own features
RelationalGCN Architecture:
- For each relation r: H^(l+1) = sum_r (A_r * H^(l) * W_r) where A_r is the adjacency for relation r
- Basis decomposition: W_r = sum_b (a_rb * B_b) with shared bases B
- Block decomposition: W_r = diag(W_r1, W_r2, …, W_rB) with smaller matrices
Key Benefits:
- Handles heterogeneous graphs with multiple edge types
- Parameter efficient through basis or block decomposition
- Captures relation-specific patterns in the data
- Effective for both entity classification and link prediction
How It Works
RelationalGCN extends Graph Convolutional Networks to handle multi-relational data where different types of edges (relations) exist between nodes.
Reference: Schlichtkrull et al., “Modeling Relational Data with Graph Convolutional Networks”, ESWC 2018.
https://arxiv.org/abs/1703.06103
Constructors
| Constructor | Summary |
|---|---|
RelationalGCNOptions() |
Initializes a new instance of the RelationalGCNOptions class with default values. |
RelationalGCNOptions(RelationalGCNOptions<T> other) |
Initializes a new instance by copying from another instance. |
Properties
| Property | Summary |
|---|---|
RelationalAggregation Aggregation { get; set; } |
Gets or sets the aggregation method for neighbor messages. |
bool UseBasisDecomposition { get; set; } |
Gets or sets whether to use basis decomposition. |
bool UseBlockDecomposition { get; set; } |
Gets or sets whether to use block decomposition. |
bool UseSelfLoop { get; set; } |
Gets or sets whether to add self-loops to the graph. |
double DropoutRate { get; set; } |
Gets or sets the dropout rate for regularization. |
double Regularization { get; set; } |
Gets or sets the regularization strength. |
int ForecastHorizon { get; set; } |
Gets or sets the forecast horizon. |
int HiddenDimension { get; set; } |
Gets or sets the hidden dimension for the model. |
int NumBases { get; set; } |
Gets or sets the number of bases for basis decomposition. |
int NumBlocks { get; set; } |
Gets or sets the number of blocks for block decomposition. |
int NumLayers { get; set; } |
Gets or sets the number of R-GCN layers. |
int NumNodes { get; set; } |
Gets or sets the number of nodes (entities) in the graph. |
int NumRelations { get; set; } |
Gets or sets the number of relation types. |
int NumSamples { get; set; } |
Gets or sets the number of samples for uncertainty estimation. |
int SequenceLength { get; set; } |
Gets or sets the sequence length (input time steps). |