The code is available here.

We additionally assume that the agent cannot interact with the environment but has access to the action-labeled transition data collected by some agents with unknown qualities.

In offline model-based optimization, we strive to maximize a black-box objective function by only leveraging a static dataset of designs and their scores. This problem setting arises in numerous fields including the design of materials, robots, DNAs, proteins, etc. Recent approaches train a deep neural network (DNN) model on the static dataset to act as a proxy function, and then perform gradient ascent on the existing designs to obtain potentially high-scoring designs. This methodology frequently suffers from the out-of-distribution problem where the proxy function often returns adversarial designs.

The Belief-Desire-Intention (BDI) model is a cornerstone for representing rational agency in artificial intelligence and cognitive sciences. Yet, its integration into structured, semantically interoperable knowledge representations remains limited. This paper presents a formal BDI Ontology, conceived as a modular Ontology Design Pattern (ODP) that captures the cognitive architecture of agents through beliefs, desires, intentions, and their dynamic interrelations. The ontology ensures semantic precision and reusability by aligning with foundational ontologies and best practices in modular design. Two complementary lines of experimentation demonstrate its applicability: (i) coupling the ontology with Large Language Models (LLMs) via Logic Augmented Generation (LAG) to assess the contribution of ontological grounding to inferential coherence and consistency; and (ii) integrating the ontology within the Semas reasoning platform, which implements the Triples-to-Beliefs-to-Triples (T2B2T) paradigm, enabling a bidirectional flow between RDF triples and agent mental states. Together, these experiments illustrate how the BDI Ontology acts as both a conceptual and operational bridge between declarative and procedural intelligence, paving the way for cognitively grounded, explainable, and semantically interoperable multi-agent and neuro-symbolic systems operating within the Web of Data.

In offline model-based optimization, we strive to maximize a black-box objective function by only leveraging a static dataset of designs and their scores. This problem setting arises in numerous fields including the design of materials, robots, DNAs, proteins, etc. Recent approaches train a deep neural network (DNN) model on the static dataset to act as a proxy function, and then perform gradient ascent on the existing designs to obtain potentially high-scoring designs. This methodology frequently suffers from the out-of-distribution problem where the proxy function often returns adversarial designs. BDI consists of two mappings: the forward mapping leverages the static dataset to predict the scores of the high-scoring designs, and the backward mapping leverages the high-scoring designs to predict the scores of the static dataset.

The origins of the BDI model lie in the theory of human practical reasoning developed by the philosopher Michael Bratman (1987) in the mid-1980s. AI Magazine Volume 22 Number 4 (2001) ( AAAI) Among the three strands of BDI research, Reasoning about Rational Agents focuses almost exclusively on the logical foundations of the theory. "If you've heard about A-life but aren't quite sure what it is or where it's going, Grand's book is an excellent place to enter one of the more exciting areas of twenty-firstcentury science." Steve Grand showed us how to build a universe of evolving creatures, without the prevailing academic biases.This delightful book is a fresh and inspiring account of how to succeed in creating artificial life." Nevertheless, it continues to be an important, thriving research area that helps to crystallize some of the deepest questions in the field.

A recurring requirement for human-level artificial intelligence is the incorporation of vast amounts of knowledge into a software agent that can use the knowledge in an efficient and organized fashion. This article discusses representations and processes for agents and behavior models that integrate large, diverse knowledge stores, are long-lived, and exhibit high degrees of competence and flexibility while interacting with complex environments. There are many different approaches to building such agents, and understanding the important commonalities and differences between approaches is often difficult. We introduce a new approach to comparing frameworks based on the notions of commitment, reconsideration, and a categorization of representations and processes. We review four agent frameworks, concentrating on the major representations and processes each directly supports.

Serious games offer an opportunity for players to learn communication skills by practicing conversations with nonplaying characters (NPCs). To realize this potential, the player needs freedom of play to discover the relationships between its actions and their effects on the partner and the conversation. Scripting is currently the common approach to design in-game dialogue. Although scripting is a robust technique, the approach tends to produce deterministic conversations, allowing little control to the player. It is claimed that a Belief-Desire-Intention (BDI) approach to model the behavior of NPCs allows greater freedom to the player, and delivers better scalability and re-use of dialogues. This claim is evaluated by using BDI in the development of a sales-talk training game in the real-estate domain. It is concluded that BDI enables representative NPCs that respond appropriately and the game allows the player its freedom of choice to explore. The results also showed that BDI brings about new challenges to address, in order to further increase the quality of in-game dialogue.

A recurring requirement for human-level artificial intelligence is the incorporation of vast amounts of knowledge into a software agent that can use the knowledge in an efficient and organized fashion. This article discusses representations and processes for agents and behavior models that integrate large, diverse knowledge stores, are long-lived, and exhibit high degrees of competence and flexibility while interacting with complex environments. There are many different approaches to building such agents, and understanding the important commonalities and differences between approaches is often difficult. We introduce a new approach to comparing frameworks based on the notions of commitment, reconsideration, and a categorization of representations and processes. We review four agent frameworks, concentrating on the major representations and processes each directly supports. By organizing the approaches according to a common nomenclature, the analysis highlights points of similarity and difference and suggests directions for integrating and unifying disparate approaches and for incorporating research results from one framework into alternatives.