Results 1  10
of
13
Map Learning with Uninterpreted Sensors and Effectors
 Artificial Intelligence
, 1997
"... This paper presents a set of methods by which a learning agent can learn a sequence of increasingly abstract and powerful interfaces to control a robot whose sensorimotor apparatus and environment are initially unknown. The result of the learning is a rich hierarchical model of the robot's world (it ..."
Abstract

Cited by 129 (20 self)
 Add to MetaCart
This paper presents a set of methods by which a learning agent can learn a sequence of increasingly abstract and powerful interfaces to control a robot whose sensorimotor apparatus and environment are initially unknown. The result of the learning is a rich hierarchical model of the robot's world (its sensorimotor apparatus and environment). The learning methods rely on generic properties of the robot's world such as almosteverywhere smooth e ects of motor control signals on sensory features. At thelowest level of the hierarchy, the learning agent analyzes the e ects of its motor control signals in order to de ne a new set of control signals, one for each of the robot's degrees of freedom. It uses a generateandtest approach to de ne sensory features that capture important aspects of the environment. It uses linear regression to learn models that characterize contextdependent e ects of the control signals on the learned features. It uses these models to de ne highlevel control laws for nding and following paths de ned using constraints on the learned features. The agent abstracts these control laws, which interact with the continuous environment, to a nite set of actions that implement discrete state transitions. At this point, the agent has abstracted the robot's continuous world to a nitestate world and can use existing methods to learn its structure. The learning agent's methods are evaluated on several simulated robots with di erent sensorimotor systems and environments.
On the Notion of Interestingness in Automated Mathematical Discovery
 International Journal of Human Computer Studies
, 2000
"... We survey ve mathematical discovery programs by looking in detail at the discovery processes they illustrate and the success they've had. We focus on how they estimate the interestingness of concepts and conjectures and extract some common notions about interestingness in automated mathematical ..."
Abstract

Cited by 64 (25 self)
 Add to MetaCart
We survey ve mathematical discovery programs by looking in detail at the discovery processes they illustrate and the success they've had. We focus on how they estimate the interestingness of concepts and conjectures and extract some common notions about interestingness in automated mathematical discovery. We detail how empirical evidence is used to give plausibility to conjectures, and the dierent ways in which a result can be thought of as novel. We also look at the ways in which the programs assess how surprising and complex a conjecture statement is, and the dierent ways in which the applicability of a concept or conjecture is used. Finally, we note how a user can set tasks for the program to achieve and how this aects the calculation of interestingness. We conclude with some hints on the use of interestingness measures for future developers of discovery programs in mathematics.
Automatic Concept Formation in Pure Mathematics
"... The HR program forms concepts and makes conjectures in domains of pure mathematics andusestheoremproverOTTERandmodel generatorMACEtoproveordisprovetheconjectures. HRmeasurespropertiesofconcepts andassessesthetheoremsandproofsinvolving themtoestimatetheinterestingnessofeach concept and employ a best ..."
Abstract

Cited by 38 (28 self)
 Add to MetaCart
The HR program forms concepts and makes conjectures in domains of pure mathematics andusestheoremproverOTTERandmodel generatorMACEtoproveordisprovetheconjectures. HRmeasurespropertiesofconcepts andassessesthetheoremsandproofsinvolving themtoestimatetheinterestingnessofeach concept and employ a best first search. This approachhasledHRtothediscoveryofinterestingnewmathematics and enables it to build theories from just the axioms of finite algebras.
Autonomous development of a grounded object ontology by a learning robot
 Proceedings of the TwentySecond Conference on Artificial Intelligence (AAAI07
, 2007
"... We describe how a physical robot can learn about objects from its own autonomous experience in the continuous world. The robot identifies statistical regularities that allow it to represent a physical object with a cluster of sensations that violate a static world model, track that cluster over time ..."
Abstract

Cited by 14 (1 self)
 Add to MetaCart
We describe how a physical robot can learn about objects from its own autonomous experience in the continuous world. The robot identifies statistical regularities that allow it to represent a physical object with a cluster of sensations that violate a static world model, track that cluster over time, extract percepts from that cluster, form concepts from similar percepts, and learn reliable actions that can be applied to objects. We present a formalism for representing the ontology for objects and actions, a learning algorithm, and the results of an evaluation with a physical robot.
Learning from the Environment Based on Percepts and Actions
, 1989
"... States Government. To the people, especially those in China. This thesis is a study of “learning from the environment. ” As machine learning moves from toy environments towards the real world, the problem of learning autonomously from environments whose structure is not completely defined a priori b ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
States Government. To the people, especially those in China. This thesis is a study of “learning from the environment. ” As machine learning moves from toy environments towards the real world, the problem of learning autonomously from environments whose structure is not completely defined a priori becomes ever more critical. Three of the major challenges are: (1) how to integrate various high level AI techniques such as exploration, problem solving, learning, and experimentation with low level perceptions and actions so that learning can be accomplished through interactions with the environment; (2) how to acquire new knowledge of the environment and to learn from mistakes autonomously, permitting incorrect information to be identified and corrected; (3) how to create features in such a way that knowledge acquisition is not limited by the initial concept description language provided by the designers of a system. The thesis defines learning from the environment as the process of inferring the laws of the environment that enable the learner to solve problems. The inputs to the learning are goals to be achieved, percepts and actions sensed and effected by the learner, and
HR  A System for Machine Discovery in Finite Algebras
 ECAI 98 Workshop Programme
, 1998
"... We describe the HR concept formation program which invents mathematical definitions and conjectures in finite algebras such as group theory and ring theory. We give the methods behind and the reasons for the concept formation in HR, an evaluation of its performance in its training domain, group theo ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
We describe the HR concept formation program which invents mathematical definitions and conjectures in finite algebras such as group theory and ring theory. We give the methods behind and the reasons for the concept formation in HR, an evaluation of its performance in its training domain, group theory, and a look at HR in domains other than group theory.
Relating Relational Learning Algorithms
 Inductive Logic Programming
, 1992
"... Relational learning algorithms are of special interest to members of the machine learning community; they offer practical methods for extending the representations used in algorithms that solve supervised learning tasks. Five approaches are currently being explored to address issues involved with us ..."
Abstract

Cited by 7 (0 self)
 Add to MetaCart
Relational learning algorithms are of special interest to members of the machine learning community; they offer practical methods for extending the representations used in algorithms that solve supervised learning tasks. Five approaches are currently being explored to address issues involved with using relational representations. This paper surveys algorithms embodying these approaches, summarizes their empirical evaluations, highlights their commonalities, and suggests potential directions for future research. Keywords: supervised learning, representation, relational learning 1 Introduction Relational learning algorithms extend the capabilities of propositional or monadic supervised learning algorithms. Supervised learning algorithms input a set of instances, which are described by a set of predictor descriptors and a target descriptor. These algorithms construct a function (i.e., a concept description) that can predict an instance's target descriptor value given its predictor desc...
Automatic Theory Formation in Graph Theory
 In Argentine Symposium on Artificial Intelligence
, 1999
"... This paper presents SCOT, a system for automatic theory construction in the domain of Graph Theory. Following on the footsteps of the programs ARE [9], HR [1] and Cyrano [6], concept discovery is modeled as search in a concept space. We propose a classification for discovery heuristics, which takes ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
This paper presents SCOT, a system for automatic theory construction in the domain of Graph Theory. Following on the footsteps of the programs ARE [9], HR [1] and Cyrano [6], concept discovery is modeled as search in a concept space. We propose a classification for discovery heuristics, which takes into account the main processes related to theory construction: concept construction, example production, example analysis, conjecture construction, and conjecture analysis.
A Computational Approach to George Boole's Discovery of Mathematical Logic
 LVO Feb 01 st
, 1997
"... This paper reports a computational model of Boole's discovery of Logic as a part of Mathematics. George Boole (18151864) found that the symbols of Logic behaved as algebraic symbols, and he then rebuilt the whole contemporary theory of Logic by the use of methods such as the solution of algebraic e ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
This paper reports a computational model of Boole's discovery of Logic as a part of Mathematics. George Boole (18151864) found that the symbols of Logic behaved as algebraic symbols, and he then rebuilt the whole contemporary theory of Logic by the use of methods such as the solution of algebraic equations. Study of the different historical factors that influenced this achievement has served as background for our two main contributions: a computational representation of Boole's Logic before it was mathematized; and a production system, BOOLE2, that rediscovers Logic as a science that behaves exactly as a branch of Mathematics, and that thus validates to some extent the historical explanation. The system's discovery methods are found to be general enough to handle three other cases: two versions of a Geometry due to a contemporary of Boole, and a small subset of the Differential Calculus. 1 Introduction In 1847, George Boole found that, by adequately representing Logic, it became a bra...
The Process of Discovery
, 1997
"... This paper argues that all discoveries, if they can be viewed as autonomous learning from the environment, share a common process. This is the process of model abstraction involving four steps: act, predict, surprise, and refine, all built on top of the discoverer's innate actions, percepts, and men ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
This paper argues that all discoveries, if they can be viewed as autonomous learning from the environment, share a common process. This is the process of model abstraction involving four steps: act, predict, surprise, and refine, all built on top of the discoverer's innate actions, percepts, and mental constructors. The evidence for this process is based on observations on various discoveries, ranging from children playing to animal discoveries of tools, from human problem solving to scientific discovery. Details of this process can be studied with computer simulations of discovery in simulated environments. 1 Introduction How is a discovery made? This is a question that interests us all. From the very beginning of civilization, philosophers have been thinking about and debating the origin of knowledge. Scientists, who search for the laws of nature, have been looking for patterns of discovery and applying them whenever possible. Psychologists, fascinated by the human mind, have been o...