William Sverdlik

Problem solving using cultural algorithms

In this paper an approach to evolutionary learning based upon principles of cultural evolution is developed. In this dual-inheritance system, there is an evolving population of trait sequences as well as an associated belief space. The belief space is derived from the behavior of individuals and is used to actively constrain the traits acquired in future populations. Shifts in the representation of the belief space and the population is supported. The approach is used to solve several versions of the BOOLE problem; F6, F11, and F20. The results are compared with other approaches and the advantages of a dual inheritance approach using cultural algorithms is discussed.<<ETX>>

Dynamic version spaces in machine learning

A hybrid learning algorithm for discovering concepts with multiple disjuncts is presented. The algorithm, HYBAL, in incorporating both version spaces and genetic algorithms, extends the work of R.G. Reynolds (1990) to learning of Boolean concepts from an exponentially growing hypothesis space. Learning is accomplished via factoring the underlying version space into tractable subspaces, and then dynamically deriving concepts for the corresponding S set and G sets. In delaying the specification of a concept language until run time, it is demonstrated that HYBAL is capable of solving a larger class of Boolean functions than with traditional version spaces, where concepts are specified at compile time.<<ETX>>

Towards long-term international collaboration in computer science education

Student success in a globally distributed organization requires effective communication skills to deal with the barriers that are inherent in such settings. This paper proposes to provide students the opportunity to be involved in collaborative “real-world” projects. International collaboration in computer science education can provide both technical and cultural benefits for all parties involved. This paper represents a work-in-progress report on the first stages of preparation to start an active long-term collaboration in computer science education between Eastern Michigan University (USA) and the University of Anbar (Iraq). We present an informal timeline for developing such collaborations and enumerate potential projects. In addition, anticipated difficulties are articulated.

Learning the behavior of Boolean circuits from examples using cultural algorithms

In this paper an approach to evolutionary learning based upon principles of cultural evolution is developed. In this dual-inheritance system, there is an evolving population of trait sequences as well as an associated belief space. The belief space is derived from the behavior of individuals and is used to actively constrain the traits acquired in future populations. Shifts in the representation of the belief space and the population are supported. The approach is used to solve several versions of the BOOLE problem; F6, F11, and F20. The results are compared with other approaches and the advantages of a dual inheritance approach using cultural algorithms is discussed.

Cluster Detection with the PYRAMID Algorithm

As databases continue to grow in size, efficient and effective clustering algorithms play a paramount role in data mining applications. Practical clustering faces several challenges including: identifying clusters of arbitrary shapes, sensitivity to the order of input, dynamic determination of the number of clusters, outlier handling, processing speed of massive data sets, handling higher dimensions, and dependence on user-supplied parameters. Many studies have addressed one or more of these challenges. PYRAMID, or parallel hybrid clustering using genetic programming and multi-objective fitness with density, is an algorithm that we introduced in a previous research, which addresses some of the above challenges. While leaving significant challenges for future work, such as handling higher dimensions, PYRAMID employs a combination of data parallelism, a form of genetic programming, and a multi-objective density-based fitness function in the context of clustering. This study adds to our previous research by exploring the detection capability of PYRAMID against a challenging dataset and evaluating its independence on user supplied parameters

Incorporating domain specific knowledge into version space search

While version spaces are useful in the conceptualization of an inductive concept learning problem, they are seldom used in practice. This is because the implementation of version spaces can use an amount of space that is exponential in terms of the amount of data presented, even for simple conjunctive learning problems. An approach is developed that uses domain knowledge to infer an associated G set hypothesis in the S set. This allows the use of G set information while concurrently restricting the space required. A prototype, the knowledge based candidate elimination (KBCE) algorithm, solves the Boole problem using fewer examples than previous approaches, and is extended to a class of Boolean functions that subsumes the multiplexor.

Hybal: A Self Tutoring Algorithm for Concept Learning in Highly Autonomous Systems

In the paper, a hybrid learning algorithm for discovering concepts with multiple disjuncts in an exponentially growing hypothesis space is presented. The approach, HYBAL, extends the work of Hirsh 141 and Reynolds [9] to produce an autonomous system that learns to partition a large search space incrementally into successively smaller search spaces using a divide and conquer strategy. This approach is used to solve the Boolean problem for a F20 multiplexor. The system needed to examine less than 0.5% of the entire search space, in order to achieve a solution.