Generative Artificial Intelligence (GenAI) lacks full alignment with human creative reasoning. In existing systems, human intention is compressed into prompts that directly target the final artifact. Thus, GenAI often falls into assumptions that violate conditions and gradually shift authority away from humans. We argue that this misalignment does not stem from GenAI model capabilities but from the absence of an explicit, persistent, machine-readable representation of human creative intention. Inspired by tracing paper in architectural design, we introduce the Rice-Paper Layer (RPL), a non-destructive, persistent, semantic layer that enables humans to externalize creative intention independently of the final artifact. We position RPL as the foundation for preserving human authority and conceptual consistency in human-AI co-creation.

We present aspects of ecosystem engineering for a strategy board game. Human and machine players of the ecosystem can pick opponents or form teams and play against other teams or players. We present the key features of the ecosystem, the highlights of the development process and we propose concrete potential uses of the ecosystem in research and education.

This paper focuses on preserving the privacy of sensitive patterns when inducing decision trees. Our record augmentation approach for hiding sensitive classification rules in binary datasets is preferred over other heuristic solutions like output perturbation or cryptographic techniques since the raw data itself is readily available for public use. We describe the process and an indicative experiment using a prototype hiding tool.

The evolution of scientific research as well as the development of the relevant technologies and methodologies in the area of distance learning fall within the scope of the objectives the Hellenic Open University whose mission is to provide distance education at both undergraduate and postgraduate level. To develop intelligent e-learning applications, we capture the semantics of the Computer Science module concepts with the use of ontologies, by following a specific methodology to build domain knowledge ontologies. We illustrate our approach by presenting the construction of the ontologies of the Artificial Intelligence & Expert Systems module.

Universities and medical companies face constantly the challenge of developing protective or adaptive countermeasures for the purpose of training new students and employees in the use of their laboratories and the assimilation of complex experimental methods. Yet the existing space and time limitations along with the high cost and sensitivity of some laboratory instruments, which are therefore not fully exposed to manipulation, result to a stiff and not always successful training procedure. The proposal deals with this problem by the creation of a game-like interactive simulation environment for the trainee to be trained in the absence of risks of undesirable consequences. The developed test-bed application is an interactive virtual environment simulating the biology laboratory at the Hellenic Open University. In this, the trainee acts by interacting with the laboratory equipment and light microscopes. By using such an application, the trainee is allowed to cause an unlimited number of unintended experimental tests and learn by "trial-and-error".

When learning how to play a strategy board game, one can measure the relative effectiveness of the learned policies by assessing how often a player wins and how easily these wins are scored. Experimental evidence also shows that when one of the competing players is trained by a sophisticated tutor, performance benefits also flow to the opponent. We present comprehensive experimental evidence that the level of tutor effectiveness is best demonstrated by the improvement of the tutored player's opponent, this performance change is termed the pendulum effect.

In this work we discuss Social Reinforcement Learning on self-trained agents. We simulate social learning by implementing a tournament on an existing board game that utilizes reinforcement learning for playing and learning. The socially trained agents are compared to self-trained agents and their superior performance is noted. The findings and the infrastructure requirements mandate the development of a Social Reinforcement Learning Multi-Agent-Based Simulation platform.

In this article we experiment with a 2-player strategy board game where playing models are developed using reinforcement learning and neural networks. The models are developed to speed up automatic game development based on human involvement at varying levels of sophistication and density when compared to fully autonomous playing. The experimental results suggest a clear and measurable association between the ability to win games and the ability to do that fast, while at the same time demonstrating that there is a minimum level of human involvement beyond which no learning really occurs.

The Hellenic Open University has embarked on a large-scale effort to enhance its textbook-based material with content that demonstrably supports the basic tenets of distance learning. The challenge is to set up a framework that allows for the production-level creation, distribution and consumption of content, and at the same time, evaluate the effort in terms of technological, educational and organizational knowledge gained. This paper presents a model of the educational process that is used as a development backbone and argues about its conceptual and technical practicality at large.

Reinforcement learning is considered as one of the most suitable and prominent methods for solving game problems due to its capability to discover good strategies by extended se self-training and limited initial knowledge. In this paper we elaborate on using reinforcement learning for verifying game designs and playing strategies. Specifically, we examine a new strategy game that has been trained on self-playing games and analyze the game performance after human interaction. We demonstrate, through selected game instances, the impact of human interference to the learning process, and eventually the game design.

We use genetic algorithms to evolve classification decision trees. The performance of the system is measured on a set of standard discretized concept learning problems and compared (very favorably) with the performance of two known algorithms (C4.5, OneR).

This work deals with stability in incremental induction of decision trees. Stability problems arise when an induction algorithm must revise a decision tree very often and oscillations between similar concepts decrease learning speed. We introduce a heuristic and an algorithm with theoretical and experimental backing to tackle this problem.