Probabilistic logic network
Encyclopedia
A probabilistic logic network (PLN) is a novel conceptual, mathematical and computational approach to uncertain inference
; inspired by logic programming
, but using probabilities in place of crisp (true/false) truth values, and fractional uncertainty in place of crisp known/unknown values
. In order to carry out effective reasoning in real-world circumstances, artificial intelligence
software must robustly handle uncertainty. However, previous approaches to uncertain inference do not have the breadth of scope required to provide an integrated treatment of the disparate forms of cognitively critical uncertainty as they manifest themselves within the various forms of pragmatic inference. Going beyond prior probabilistic approaches to uncertain inference, PLN is able to encompass within uncertain logic such ideas as induction, abduction, analogy, fuzziness and speculation, and reasoning about time and causality.
PLN was developed by Ben Goertzel
, Matt Ikle, Izabela Lyon Freire Goertzel and Ari Heljakka for use as a cognitive algorithm used by MindAgents within the OpenCog
Core. PLN was developed originally for use within the Novamente Cognition Engine.
and predicate logic
, and scales up to operate in real time on large dynamic knowledge bases.
The goal underlying the theoretical development of PLN has been the creation of practical software systems carrying out complex, useful inferences based on uncertain knowledge and drawing uncertain conclusions. PLN has been designed to allow basic probabilistic inference to interact with other kinds of inference such as intensional
inference, fuzzy inference
, and higher-order inference using quantifiers, variables, and combinators, and be a more convenient approach than Bayesian networks (or other conventional approaches) for the purpose of interfacing basic probabilistic inference with these other sorts of inference.
and combinatory
logic, as well as some aspects of predicate logic and autoepistemic logic
, to form a complete inference system, tailored for easy integration with software components embodying other (not explicitly logical) aspects of intelligence.
PLN represents truth values as intervals, but with different semantics than in Imprecise Probability Theory. In addition to the interpretation of truth in a probabilistic fashion, a truth value in PLN also has an associated amount of certainty. This generalizes the notion of truth values used in autoepistemic logic
, where truth values are either known or unknown, and when known, are either true or false.
The current version of PLN has been used in narrow-AI applications such as the inference of biological hypotheses from knowledge extracted from biological texts via language processing, and to assist the reinforcement learning
of an embodied agent, in a simple virtual world
, as it is taught to play "fetch".
Uncertain inference
Uncertain inference was first described by Rijsbergen as a way to formally define a query and document relationship in Information retrieval. This formalization is a logical implication with an attached measure of uncertainty.-Definitions:...
; inspired by logic programming
Logic programming
Logic programming is, in its broadest sense, the use of mathematical logic for computer programming. In this view of logic programming, which can be traced at least as far back as John McCarthy's [1958] advice-taker proposal, logic is used as a purely declarative representation language, and a...
, but using probabilities in place of crisp (true/false) truth values, and fractional uncertainty in place of crisp known/unknown values
Autoepistemic logic
The autoepistemic logic is a formal logic for the representation and reasoning of knowledge about knowledge. While propositional logic can only express facts, autoepistemic logic can express knowledge and lack of knowledge about facts....
. In order to carry out effective reasoning in real-world circumstances, artificial intelligence
Artificial intelligence
Artificial intelligence is the intelligence of machines and the branch of computer science that aims to create it. AI textbooks define the field as "the study and design of intelligent agents" where an intelligent agent is a system that perceives its environment and takes actions that maximize its...
software must robustly handle uncertainty. However, previous approaches to uncertain inference do not have the breadth of scope required to provide an integrated treatment of the disparate forms of cognitively critical uncertainty as they manifest themselves within the various forms of pragmatic inference. Going beyond prior probabilistic approaches to uncertain inference, PLN is able to encompass within uncertain logic such ideas as induction, abduction, analogy, fuzziness and speculation, and reasoning about time and causality.
PLN was developed by Ben Goertzel
Ben Goertzel
Ben Goertzel , is an American author and researcher in the field of artificial intelligence. He currently leads Novamente LLC, a privately held software company that attempts to develop a form of strong AI, which he calls "Artificial General Intelligence"...
, Matt Ikle, Izabela Lyon Freire Goertzel and Ari Heljakka for use as a cognitive algorithm used by MindAgents within the OpenCog
OpenCog
OpenCog is a project that aims to build an open source artificial general intelligence framework. OpenCog Prime is a specific set of interacting components designed to give rise to human-equivalent artificial general intelligence...
Core. PLN was developed originally for use within the Novamente Cognition Engine.
Goal
The basic goal of PLN is to provide reasonably accurate probabilistic inference in a way that is compatible with both term logicTerm logic
In philosophy, term logic, also known as traditional logic or aristotelian logic, is a loose name for the way of doing logic that began with Aristotle and that was dominant until the advent of modern predicate logic in the late nineteenth century...
and predicate logic
Predicate logic
In mathematical logic, predicate logic is the generic term for symbolic formal systems like first-order logic, second-order logic, many-sorted logic or infinitary logic. This formal system is distinguished from other systems in that its formulae contain variables which can be quantified...
, and scales up to operate in real time on large dynamic knowledge bases.
The goal underlying the theoretical development of PLN has been the creation of practical software systems carrying out complex, useful inferences based on uncertain knowledge and drawing uncertain conclusions. PLN has been designed to allow basic probabilistic inference to interact with other kinds of inference such as intensional
Intensional
Intensional* in philosophy of language: not extensional. See also intensional definition versus extensional definition.* in philosophy of mind: an intensional state is a state which has a propositional content....
inference, fuzzy inference
Fuzzy logic
Fuzzy logic is a form of many-valued logic; it deals with reasoning that is approximate rather than fixed and exact. In contrast with traditional logic theory, where binary sets have two-valued logic: true or false, fuzzy logic variables may have a truth value that ranges in degree between 0 and 1...
, and higher-order inference using quantifiers, variables, and combinators, and be a more convenient approach than Bayesian networks (or other conventional approaches) for the purpose of interfacing basic probabilistic inference with these other sorts of inference.
Implementation
PLN begins with a term logic foundation, and then adds on elements of probabilisticProbabilistic logic
The aim of a probabilistic logic is to combine the capacity of probability theory to handle uncertainty with the capacity of deductive logic to exploit structure. The result is a richer and more expressive formalism with a broad range of possible application areas...
and combinatory
Combinatory logic
Combinatory logic is a notation introduced by Moses Schönfinkel and Haskell Curry to eliminate the need for variables in mathematical logic. It has more recently been used in computer science as a theoretical model of computation and also as a basis for the design of functional programming...
logic, as well as some aspects of predicate logic and autoepistemic logic
Autoepistemic logic
The autoepistemic logic is a formal logic for the representation and reasoning of knowledge about knowledge. While propositional logic can only express facts, autoepistemic logic can express knowledge and lack of knowledge about facts....
, to form a complete inference system, tailored for easy integration with software components embodying other (not explicitly logical) aspects of intelligence.
PLN represents truth values as intervals, but with different semantics than in Imprecise Probability Theory. In addition to the interpretation of truth in a probabilistic fashion, a truth value in PLN also has an associated amount of certainty. This generalizes the notion of truth values used in autoepistemic logic
Autoepistemic logic
The autoepistemic logic is a formal logic for the representation and reasoning of knowledge about knowledge. While propositional logic can only express facts, autoepistemic logic can express knowledge and lack of knowledge about facts....
, where truth values are either known or unknown, and when known, are either true or false.
The current version of PLN has been used in narrow-AI applications such as the inference of biological hypotheses from knowledge extracted from biological texts via language processing, and to assist the reinforcement learning
Reinforcement learning
Inspired by behaviorist psychology, reinforcement learning is an area of machine learning in computer science, concerned with how an agent ought to take actions in an environment so as to maximize some notion of cumulative reward...
of an embodied agent, in a simple virtual world
Virtual world
A virtual world is an online community that takes the form of a computer-based simulated environment through which users can interact with one another and use and create objects. The term has become largely synonymous with interactive 3D virtual environments, where the users take the form of...
, as it is taught to play "fetch".