
Improving the Knowledge-Based Expert System Lifecycle
... Modern experts no longer predominantly use pencil and paper to solve problems. Instead, they use some form of software application created to help them work through the problem and create a solution. The process is still manual because all knowledge still resides with the experts. However, the solut ...
... Modern experts no longer predominantly use pencil and paper to solve problems. Instead, they use some form of software application created to help them work through the problem and create a solution. The process is still manual because all knowledge still resides with the experts. However, the solut ...
Relational Dynamic Bayesian Networks
... In this paper we introduce relational dynamic Bayesian networks (RDBNs) which extend DBNs to first-order (relational) domains. RDBNs subsume DPRMs and have several advantages over them, including greater simplicity and expressivity. Furthermore, they may be more easily learned using ILP techniques. ...
... In this paper we introduce relational dynamic Bayesian networks (RDBNs) which extend DBNs to first-order (relational) domains. RDBNs subsume DPRMs and have several advantages over them, including greater simplicity and expressivity. Furthermore, they may be more easily learned using ILP techniques. ...
A Market-Based Study of Optimal ATM`S Deployment Strategy
... The problem of ATM deployment is seen to be NP-complete problem (it is analogous to the file server placement problem) [3]. In order to solve this problem, three algorithms are designed and compared namely; Heuristic Approach using Convolution (HAC) [4], Rank Based Genetic Algorithm using Convolutio ...
... The problem of ATM deployment is seen to be NP-complete problem (it is analogous to the file server placement problem) [3]. In order to solve this problem, three algorithms are designed and compared namely; Heuristic Approach using Convolution (HAC) [4], Rank Based Genetic Algorithm using Convolutio ...
Reinforcement Learning and Automated Planning
... Usually, in the description of domains, action schemas (also called operators) are used instead of actions. Action schemas contain variables that can be instantiated using the available objects and this makes the encoding of the domain easier. The choice of the language in which the planning problem ...
... Usually, in the description of domains, action schemas (also called operators) are used instead of actions. Action schemas contain variables that can be instantiated using the available objects and this makes the encoding of the domain easier. The choice of the language in which the planning problem ...
GA-FreeCell: Evolving Solvers for the Game of FreeCell
... We will show that not only do we solve 98% of the Microsoft 32K problem set, a result far better than the best solver on record, but we also do so significantly more efficiently in terms of time to solve, space (number of nodes expanded), and solution length (number of nodes along the path to the co ...
... We will show that not only do we solve 98% of the Microsoft 32K problem set, a result far better than the best solver on record, but we also do so significantly more efficiently in terms of time to solve, space (number of nodes expanded), and solution length (number of nodes along the path to the co ...
The Specification of Agent Behavior by Ordinary People: A Case Study
... E-Agent: via a pre-defined variable (e.g., Bring.last(),Bring.$x$.count()) or, less commonly, by utilizing an explicit SQL query over a virtual table constructed from the RDF (e.g., as with TotalGuests). The former method is more convenient and allows the author to easily specify decisions based on ...
... E-Agent: via a pre-defined variable (e.g., Bring.last(),Bring.$x$.count()) or, less commonly, by utilizing an explicit SQL query over a virtual table constructed from the RDF (e.g., as with TotalGuests). The former method is more convenient and allows the author to easily specify decisions based on ...
Average Convergence Rate of Evolutionary Algorithms
... matrix associated with an EA. A lower bound of convergence rate is derived in [2] for simple genetic algorithms by analyzing eigenvalues of the transition matrix. Then the work is extended in [3] and it is found that the convergence rate is determined by the second largest eigenvalue of the transiti ...
... matrix associated with an EA. A lower bound of convergence rate is derived in [2] for simple genetic algorithms by analyzing eigenvalues of the transition matrix. Then the work is extended in [3] and it is found that the convergence rate is determined by the second largest eigenvalue of the transiti ...
Software Reuse for Mobile Robot Applications Through Analysis
... used across domains at early level of development. Developing software for a mobile robot system involves multi-disciplines expert knowledge which includes embedded systems, real-time software issues, control theories and artificial intelligence aspects. This paper focuses on analysis patterns as a ...
... used across domains at early level of development. Developing software for a mobile robot system involves multi-disciplines expert knowledge which includes embedded systems, real-time software issues, control theories and artificial intelligence aspects. This paper focuses on analysis patterns as a ...
Learning Action Models for Multi-Agent Planning
... system [11]. In the past, there have been several works on learning action models for single agents, such as ARMS [16] and SLAF [1]. However, these learning algorithms did not take into account multiagent situations. One possibility in tackling this multi-agent learning problem is to assume that the ...
... system [11]. In the past, there have been several works on learning action models for single agents, such as ARMS [16] and SLAF [1]. However, these learning algorithms did not take into account multiagent situations. One possibility in tackling this multi-agent learning problem is to assume that the ...
A Concentration Inequalities B Benchmark C
... Step 1: Bound |OPT − OPT|. ˆ 2γ − OPTγ | Step 2: Bound |OPT Step 1 bound can be borrowed from the work on Online Stochastic Convex Programming in [4]: since µ∗ , W ∗ is known, so there is effectively full information before making the decision, i.e., consider ...
... Step 1: Bound |OPT − OPT|. ˆ 2γ − OPTγ | Step 2: Bound |OPT Step 1 bound can be borrowed from the work on Online Stochastic Convex Programming in [4]: since µ∗ , W ∗ is known, so there is effectively full information before making the decision, i.e., consider ...
[pdf]
... see this, suppose there is a decision tree of the same structure as in Figure 1, but has depth d. According to Theorem 2, it can be represented exactly with Fourier coefficients up to degree d. In this specific example, the number of non-zero Fourier coefficients is O(22d ). Nonetheless, no two vari ...
... see this, suppose there is a decision tree of the same structure as in Figure 1, but has depth d. According to Theorem 2, it can be represented exactly with Fourier coefficients up to degree d. In this specific example, the number of non-zero Fourier coefficients is O(22d ). Nonetheless, no two vari ...
A Low-Cost Approximate Minimal Hitting Set Algorithm
... comes component 1. As it is not involved in all sets, it is combined with those components that are involved in all sets except the ones already covered by 1 (note, that combinations involving 3 are no longer considered due to subsumption). This would lead us to find {1,2} as a second minimal hittin ...
... comes component 1. As it is not involved in all sets, it is combined with those components that are involved in all sets except the ones already covered by 1 (note, that combinations involving 3 are no longer considered due to subsumption). This would lead us to find {1,2} as a second minimal hittin ...
The Quest for Efficient Boolean Satisfiability Solvers | SpringerLink
... For the efficiency of the solver, the propositional formula instance is usually presented in a Product of Sum form, usually called a Conjunctive Normal Form (CNF). It is not a limitation to require the instance to be presented in CNF. There exist polynomial algorithms (e.g. [24]) to transform any pr ...
... For the efficiency of the solver, the propositional formula instance is usually presented in a Product of Sum form, usually called a Conjunctive Normal Form (CNF). It is not a limitation to require the instance to be presented in CNF. There exist polynomial algorithms (e.g. [24]) to transform any pr ...
TSTP Data-Exchange Formats for Automated Theorem Proving Tools
... As ATP systems move into real application areas, they are typically embedded as just one component in a larger system, including tools for proof analysis, transformation, presentation, and verification. In this environment, the output from one tool is often used as input to another. It is therefore ...
... As ATP systems move into real application areas, they are typically embedded as just one component in a larger system, including tools for proof analysis, transformation, presentation, and verification. In this environment, the output from one tool is often used as input to another. It is therefore ...
A Computational Intelligence Approach to Modelling Interstate Conflict
... study of interstate conflict has been the adoption of the generic term of “conflict” rather than “war” or “dispute”. This has led to collection of MID data which allows us, not only to concentrate on intense state interactions, but also on sub war interactions, where militarised behaviour occurs wit ...
... study of interstate conflict has been the adoption of the generic term of “conflict” rather than “war” or “dispute”. This has led to collection of MID data which allows us, not only to concentrate on intense state interactions, but also on sub war interactions, where militarised behaviour occurs wit ...