User:CCLevy/Probabilistic soft logic
 | |
| Developer(s) | Google Brain Team[1] |
|---|---|
| Initial release | November 9, 2015 |
| Stable release | 2.2.2[2]
/ May 6, 2020 |
| Repository | github |
| Written in | Python, C++, CUDA |
| Platform | Linux, macOS, Windows, Android, JavaScript[3] |
| Type | Machine learning library |
| License | Apache License 2.0 |
| Website | www |
Probabilistic soft logic (PSL) is a SRL framework for collective, probabilistic reasoning in relational domains. PSL uses first order logic rules as a template language for graphical models over random variables with soft truth values from the interval [0,1].[4]
Description[edit]
In recent years there has been a rise in the approaches that combine graphical models and first-order logic to allow the development of complex probabilistic models with relational structures. A notable example of such approaches is Markov logic networks (MLNs).[5] Like MLNs PSL is a modelling language (with an accompanying implementation[6]) for learning and predicting in relational domains. Unlike MLNs, PSL uses soft truth values for predicates in an interval between [0,1]. This allows for the underlying inference to be solved quickly as a convex optimization problem. This is useful in problems such as collective classification, link prediction, social network modelling, and object identification/entity resolution/record linkage.
See also[edit]
References[edit]
- ^ Cite error: The named reference
Creditswas invoked but never defined (see the help page). - ^ "PSL".
- ^ Cite error: The named reference
jswas invoked but never defined (see the help page). - ^ Bach, Stephen; Broecheler, Matthias; Huang, Bert; Getoor, Lise (2017). "Hinge-Loss Markov Random Fields and Probabilistic Soft Logic". Journal of Machine Learning Research. 18: 1–67.
- ^ Getoor, Lise; Taskar, Ben (October 12, 2007). Introduction to Statistical Relational Learning. MIT Press. ISBN 0262072882.
- ^ "GitHub repository". Retrieved March 26, 2018.