Showing content from https://github.com/opcode81/ProbCog/wiki/Features below:
Features · opcode81/ProbCog Wiki · GitHub
Skip to content Navigation Menu
Search code, repositories, users, issues, pull requests...
Saved searches Use saved searches to filter your results more quickly
Sign up You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert
opcode81 edited this page
May 9, 2013
·
1 revision Statistical Relational Models Bayesian Logic Networks (BLNs) [Java]
- Inference (posterior marginals)
- Sampling Methods
- Likelihood Weighting
- Backward Simulation
- SampleSearch
- Message-Passing Methods
- Iterative join-graph propagation (IJGP)
- (Loopy) Belief Propagation
- Exact Methods
- ACE
- Enumeration-Ask
- Variable Elimination
- Pearl's Algorithm (with join-tree clustering)
- Markov Chain Monte Carlo Methods
- ...and quite a few more
- Learning (maximum likelihood)
- Translation to MLNs
- Graphical Learning and Inference Tools
Adaptive Markov Logic Networks (AMLNs) [Python]
- Inference (posterior marginal probabilties of formulas)
- MC-SAT
- Gibbs Sampling
- enumeration (exact)
- Probabilistic Inference with Uncertain Evidence (Soft Evidential Update)
- MC-SAT-PC
- IPFP-M (iterative proportional fitting)
- Inference (most probable explanation)
- Learning
- maximum pseudo-likelihood
- maximum likelihood
- Knowledge Representation
- Cardinality restrictions (count constraints)
- Constraints on (prior) marginal probabilities of formulas
- Constraints on posterior probabilities of ground atoms and formulas (soft evidence)
Markov Logic Networks [Java]
- Inference (posterior marginals of ground atoms)
- Inference (most probable explanation)
- MaxWalkSAT
- Toulbar2 Branch&Bound
- Translation to WCSPs (weighted contraint satisfaction problems)
Unified Interfaces to Relational Models and Data
- Model Abstraction: unified interface to statistical relational models
- in the probcog Java library (and the matching jyprobcog Jython scripting library)
- in a YARP service
- in a ROS node
- Data Collection
- srldb library for working with relational data (Java/Jython); features:
- automatic generation of training databases in various formats
- automatic discretization of continuous data
- generation of basic MLN/BLN models (containing all the necessary declarations) as a starting point for learning problems
- Generation of synthetic relational data: library for convenient scripting of relational stochastic processes
Probabilistic Graphical Models
- Support for various file formats
- import: Bayesian Interchange Format (XML-BIF), Ergo/UAI, PMML 3 (extended)
- export: Bayesian Interchange Format (XML-BIF), Ergo/UAI, PMML 3 (extended), Hugin
- Inference
- Sampling Methods
- Likelihood Weighting
- Backward Simulation
- SampleSearch
- Message-Passing Methods
- Iterative join-graph propagation (IJGP)
- (Loopy) Belief Propagation
- Exact Methods
- ACE
- Enumeration-Ask
- Variable Elimination
- Pearl's Algorithm
- Markov Chain Monte Carlo Methods
- ...and quite a few more
- Learning (maximum likelihood)
Our implementation is based on BNJ (Bayesian Network Tools in Java).
Toggle table of contents Pages 22
Clone this wiki locally
You can’t perform that action at this time.
RetroSearch is an open source project built by @garambo
| Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.3