Orpheus: Engineering Multiagent Systems via Communicating Agents
Authors: Matteo Baldoni, Samuel H. Christie V, Munindar P. Singh, Amit K. Chopra
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | We propose Orpheus, a novel programming model for communicating agents based on information protocols and realized using cognitive programming. Whereas traditional models are focused on reactions to handle incoming messages, Orpheus supports organizing the internal logic of an agent based on its goals. We give an operational semantics for Orpheus and implement this semantics in an adapter to help build agents. We use the adapter to demonstrate how Orpheus simplifies the programming of decentralized multiagent systems compared to the reactive programming model. Section 6 describes a conceptual evaluation of our approach, showing key distinctions and how it meets important criteria. |
| Researcher Affiliation | Academia | 1Dipartimento di Informatica, Universit a degli Studi di Torino, Corso Svizzera 185, 10149 Torino, Italy 2Department of Computer Science, North Carolina State University, Raleigh, NC 27695, USA 3School of Computing and Communications, Lancaster University, Lancaster LA1 4WA, UK |
| Pseudocode | Yes | Listing 1: Jason snippet of a MERCHANT agent. Listing 2: Initial Net Bill Protocol (goods before pay) Listing 3: Plan pattern and Orpheus primitives. Listing 4: Generated Jason code for computing enabled quotes; the code generated for other messages is similar. Listing 5: The programmer identifies MERCHANT agent s goals and what messages these goals involve sending, and also specifies complete (not shown). Listing 6: Flexible Net Bill. Listing 7: Modified plan for !send goods. Listing 8: Fragment of Logistics relevant to PACKER. Listing 9: Plan in Orpheus for sending packed messages. The correlation of label with item is automatic. |
| Open Source Code | Yes | Code https://gitlab.com/masr |
| Open Datasets | No | The paper uses 'Net Bill (Sirbu 1997)' and a 'logistics scenario (Sicari, Rizzardi, and Coen Porisini 2019)' as running examples. These are conceptual scenarios and protocols used for demonstration, not datasets for empirical experiments. No specific datasets with access information for experimental evaluation are mentioned. |
| Dataset Splits | No | The paper does not describe any empirical experiments using datasets, thus there are no mentions of training/test/validation dataset splits. |
| Hardware Specification | No | The paper describes a programming model and its conceptual evaluation, focusing on qualitative comparisons and code structure. It does not mention any specific hardware (GPUs, CPUs, etc.) used for running experiments or evaluating performance. |
| Software Dependencies | No | The paper mentions Jason as an example BDI approach and states an implementation in Jason, but does not provide specific version numbers for Jason or any other software libraries or platforms used for their implementation. For example: 'We adopt Jason as an exemplar BDI approach...' and '...implementation in Jason.' |
| Experiment Setup | No | The paper presents a conceptual evaluation of a programming model, demonstrating its features through code examples and qualitative comparisons rather than empirical experiments. Consequently, there are no specific experimental setup details such as hyperparameters, training configurations, or system-level settings mentioned. |