Klára Pešková

What does your actor remember? towards characters with a full episodic memory

A typical present-day virtual actor is able to store episodes in an ad hoc manner, which does not allow for reconstructing the actors personal stories. This paper proposes a virtual RPG actor with a full episodic memory, which allows for this reconstruction. The paper presents the memory architecture, overviews the prototype implementation, presents a benchmark for the efficiency of the memory measurement, and details the conducted tests.

Meta Learning in Multi-agent Systems for Data Mining

In this paper we present the Pikater multi-agent system designed for solving complex data mining tasks. We emphasize the unique intelligent features of the system -- its ability to search the parameter space of the data mining methods to find the optimal configuration, and meta learning -- finding the best possible method for the given data based on the ontological compatibility of datasets.

Towards Characters with a Full Episodic Memory

A typical present-day virtual actor is able to store episodes in an ad hocmanner, which does not allow for reconstructing the actors personal stories. We have prototyped a virtual RPG actor with a fullepisodic memory, which allows for this reconstruction. The paper overviews the work done and sketches the work in progress.

Where did i put my glasses?: determining trustfulness of records in episodic memory by means of an associative network

Episodic memory represents personal history of an entity. Humanlike agents with a full episodic memory are able to reconstruct their personal stories to a large extent. Since these agents typically live in dynamic environments that change beyond their capabilities, their memory must cope with determining trustfulness of memory records. In this paper, we propose an associative network addressing this issue with regard to records about objects an agent met during its live. The network is presently being implemented into our case-study human-like agent with a full episodic memory.

Co-evolutionary genetic programming for dataset similarity induction

Metalearning deals with an important problem in machine-learning, namely selecting the right techniques to model the data at hand. In most of the metalearning approaches, a notion of similarity between datasets is needed. Our approach derives the similarity measure by combining arbitrary attribute similarity functions ordered by the optimal attribute assignment. In this paper, we propose a genetic programming based approach to the evolution of an attribute similarity inducing function. The function is composed of two parts - one describes the similarity of categorical attributes, the other describes the similarity of numerical attributes. Co-evolution is used to put these two parts together to form the similarity function. We use a repairing approach to guarantee some of the metric features for this function, and also discuss which of these features are important in metalearning.

A Novel Meta Learning System and Its Application to Optimization of Computing Agents' Results

We present a description of our multi-agent system where computational intelligence methods are embodied as software agents. This system is designed in order to allow easy experiments with learning, meta learning, gathering experience based on previous computations, and recommending suitable methods for particular data. The architecture of the system is presented and its meta learning abilities are demonstrated on a set of experiments with neural network models and both evolutionary and local search heuristics.