Rem W. Collier

EUMAS Conference 2025 Conference Paper

Agency and Generation: Friends or Enemies?

• Daniela Briola
• Rem W. Collier
• Angelo Ferrando
• Viviana Mascardi

EUMAS Conference 2025 Conference Paper

Agent Toolkits: Towards Explainable Agency in ASTRA

• Katharine Beaumont
• Elena Yan
• Samuele Burattini
• Rem W. Collier

EUMAS Conference 2025 Conference Paper

AgentSpeak(AI): Deliberate Learning Strategies for Plan Acquisition in BDI Agents

• Katharine Beaumont
• Rem W. Collier

EUMAS Conference 2025 Conference Paper

astra-langchain4j: Experiences Combining LLMs and Agent Programming

• Rem W. Collier
• Katharine Beaumont
• Andrei Ciortea

EUMAS Conference 2025 Conference Paper

Towards a Distinct Programming Style for AgentSpeak(L)

• Rem W. Collier
• Katharine Beaumont
• Andrei Ciortea

EUMAS Conference 2024 Conference Paper

CommonHealth: Multi-agent Evaluation of Blockchain-Based Patient-Centred Health Networks

• Mark Matthews
• Rem W. Collier
• Evan Spendlove

Abstract Information is essential to delivering effective healthcare. Health information traditionally resides with the entity providing healthcare to the patient. However, this is just one model for how information can be situated within a healthcare context. Decentralized health networks provide the infrastructure to realize a different information and computational model that centers the locus of health information around the patients. In this paper, we present a framework for building and testing decentralized health networks that enable patients to share their data with a communal network to gain value derived from aggregated data. While decentralized networks disperse control of health information among a network of individuals, they require careful design and testing to ensure the network’s incentives promote the interests of each member, scale cost-effectively, and protect privacy. Deploying a decentralized health network without appropriate testing presents multiple risks, including the risk of data breach, erroneous machine learning model predictions, and costly platform management. To counter these risks, we demonstrate how multi-agent-based evaluation can be central to testing and refining patient-centered decentralized health networks such that the incentives are robust, sustainable, and less susceptible to exploitation. We find that agent-based approaches can help speed up development and build confidence in scalability, safety, and effective functioning prior to real-world deployment. We also provide design recommendations and suggestions for future work.

EUMAS Conference 2024 Conference Paper

Do You Want to Play a Game? Learning to Play Tic-Tac-Toe in Hypermedia Environments

• Katharine Beaumont
• Rem W. Collier

Abstract We demonstrate the integration of Transfer Learning into a hypermedia Multi-Agent System using the Multi-Agent MicroServices (MAMS) architectural style. Agents use RDF knowledge stores to reason over information and apply Reinforcement Learning techniques to learn how to interact with a Tic-Tac-Toe API. Agents form advisor-advisee relationships in order to speed up individual learning and exploit and learn from data on the Web.