HC (bnlearn)

Module name

bnlearn_hc

Package

bnlearn

Version

4.8.3

Language

R

Docs

here

Paper

Scutari et al.[1], Norvig[2]

Graph type

DAG

MCMC

No

Edge constraints

Yes

Data type

C, D, M

Data missingness

Intervention type

Docker

bpimages/bnlearn:4.8.3

Hill-Climbing

Hill climbing (HC) is a score-based algorithm which starts with a DAG with no edges and adds, deletes or reverses edges in a greedy manner until an optimum is reached.

Some fields described

  • edgeConstraints Name of the JSON file containing background knowledge

Example JSON

[
  {
    "id": "hc-bde",
    "score": "bde",
    "iss": [
      0.001,
      0.01,
      0.1
    ],
    "issmu": null,
    "issw": null,
    "l": 5,
    "k": 1,
    "prior": "uniform",
    "beta": 1,
    "restart": 0,
    "perturb": 1,
    "timeout": null,
    "edgeConstraints": "edgeConstraints.json"
  },
  {
    "id": "hc-bge",
    "score": "bge",
    "iss": 1,
    "issmu": [
      0.0001,
      0.001,
      0.01,
      0.05
    ],
    "issw": null,
    "l": 5,
    "k": 1,
    "prior": "uniform",
    "beta": 1,
    "restart": 0,
    "perturb": 1,
    "timeout": null,
    "edgeConstraints": "edgeConstraints.json"
  }
]