Physics is like sex: sure, it may give some practical results, but that's not why we do it.
Richard Feynman
(The same could be said about science)

How to do Research

Although I have accumulated some experience in doing research and how to get papers out, I do not claim to have such a vast experience that I can really prescribe others to follow me. At any rate, much more experienced people have written very good guides on how to do good research, and below I link the guides that have served me well during my PhD and after:

Besides doing good research, one must also do good reporting of said research. Good writing skills are essential not only for the budding scientist, but for anyone whose ideas must be communicated to others. Therefore, if you want to be a leader and be taken seriously, you must write well. In order to address this I highly recommend reading William Strunk's, The Elements of Style. The book might be a little dated, but the advice stands the test of time, in particular regarding style.

Interests

My research interests span a wide variety of applications of Artificial Intelligence, primarily in the following areas:

Multiagent Systems

Agent-based software is an established method for modelling an increasingly important number of network-centred systems. Unlike traditional object-oriented approaches to system modelling, agents are able to control their own internal state and behaviour, and have dynamic relationships among themselves and the environment in which they operate. Until recently, these agent properties have been largely explored on the theoretical level and many practical questions need to be addressed before agent-oriented programming becomes an everyday reality. Thus, practical agent programming languages make up a rapidly developing subfield within agent systems research, yet there are still many unresolved problems.

Normative Reasoning

Systems of autonomous and self-interested agents interacting to achieve individual and collective goals may exhibit undesirable or unexpected properties if left unconstrained. This has led to efforts in defining systems of norms, effectively formalised laws, in terms of obligations, permissions and prohibitions. However, only relatively recently has there been efforts in modifying agent reasoning mechanisms to process norms so that an agent can deliberately, as opposed to by design (or lack of mechanisms), comply with or violate norms. In this context, I contributed to the CONTRACT project, designing and implementing a number of agent reasoning mechanisms to process norms in the context of the aerospace engine care market. My work in this area focuses on designing algorithms for norm processing and integrating them into agent reasoning mechanisms. These efforts have yielded the following publications:

MENEGUZZI, Felipe; MODGIL, Sanjay; OREN, Nir; MILES, Simon; LUCK, Michael; FACI, Noura; Applying electronic contracting to the aerospace aftercare domain, Engineering Applications of Artificial Intelligence, Available online 8 July 2012.
[DOI] [BibTeX]
OH, Jean; MENEGUZZI, Felipe; SYCARA, Katia and NORMAN, Timothy J. Introduction to Prognostic Normative Reasoning, in Lecture Notes in Computer Science 7068, 2012.
[PDF] [BibTeX]
OREN, Nir; VASCONCELOS, Wamberto; MENEGUZZI, Felipe and LUCK, Michael. Acting on Norm Constrained Plans, in Computational Logic in Multi-Agent Systems (CLIMA XII), Barcelona, Spain, 2011.
[PDF] [BibTeX] [Slides]
MILES, Simon; OREN, Nir; LUCK, Michael; MODGIL, Sanjay; MENEGUZZI, Felipe; FACI, Nora; HOLT, Camden; VICKERS, Gary. Electronic Business Contracts between Services, in Handbook of Research on P2P and Grid Systems for Service-Oriented Computing: Models, Methodologies and Applications, published by IGI Global, 2010.
[PDF] [BibTeX]
MENEGUZZI, Felipe; OREN, Nir and VASCONCELOS, Wamberto. Using constraints for Norm-aware BDI Agents, in The Fourth Annual Conference of the International Technology Alliance, London, UK, 2010.
[PDF] [BibTeX]
MENEGUZZI, Felipe; MODGIL, Sanjay; OREN, Nir; MILES, Simon; LUCK Michael; HOLT, Camden; SMITH, Malcolm. A contract-based system for aerospace aftercare, in Workshop on Technological trends in Enterprise Systems for SMEs and large enterprises: Heading towards the Future Internet, London, UK, 2009.
[PDF] [BibTeX] [Slides]
MENEGUZZI, Felipe and LUCK, Michael Norm-based behaviour modification in BDI agents. Proceedings of the 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Budapest, Hungary, 2009.
[PDF] [BibTeX] [Slides]
MODGIL, Sanjay; FACI, Noura; MENEGUZZI, Felipe; OREN, Nir; MILES, Simon; LUCK, Michael A Framework for Monitoring Agent-Based Normative Systems. Proceedings of the 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Budapest, Hungary, 2009.
[PDF] [BibTeX] [Slides]
MENEGUZZI, Felipe; MODGIL, Sanjay; OREN, Nir; MILES, Simon; LUCK Michael; FACI, Noura; HOLT, Camden; SMITH, Malcolm. Monitoring and Explanation of Contract Execution: A Case Study in the Aerospace Domain . Proceedings of the 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS) - Industry and Applications Track, Budapest, Hungary, 2009.
[PDF] [BibTeX] [Slides]

Agent Reasoning and the BDI Model

One of the most widely studied agent reasoning models is based on the notions of beliefs, desires and intentions (or BDI) as mental attitudes that guide the selection of courses of action. In this model, beliefs describe knowledge about the world, while desires are states of affairs to achieve, and intentions are commitments to achieving a particular subset of desires. The BDI model has its origins in the philosophical work of Bratman to explain the way in which humans select a series of actions directed at the achievement of a larger goal while avoiding spending time considering less important ones. Most of my work on agent reasoning draws heavily on this particular model either directly, in my work on agent languages, or indirectly, in the architectures for proactive assistance I worked on. My research in this area has yielded the following publications:

MENEGUZZI, Felipe. Motivations and Goal-Directed Autonomy, in AAAI-10 Workshop on Goal-Directed Autonomy, 2010 (invited paper).
[PDF] [BibTeX] [Slides]
MENEGUZZI, Felipe Extending agent languages for autonomy. AAMAS 2008 Doctoral Mentoring Programme. Estoril, Portugal, 2008.
[PDF] [BibTeX] [Slides]
MENEGUZZI, F.; LUCK, Michael; Interaction among agents that plan. From Agent Theory to Agent Implementation. Estoril, Portugal, 2008.
[PDF] [BibTeX] [Slides]
MENEGUZZI, F. and LUCK, M.. Motivations as an abstraction of meta-level reasoning. Proceedings of the 5th International Central and Eastern European Conference on Multi-Agent Systems. Leipzig, 2007.
[PDF] [BibTeX] [Slides]

Agent Programming Languages

One major push of my research during my PhD was focused on creating agent languages and architectures that overcome these limitations by the integration of AI techniques within the underlying agent interpreters. In my thesis I created AgentSpeak(PL), an extension of a traditional logic-based BDI-style agent language, which includes a mechanism for processing goals in a manner that decouples goal achievement from plan execution, as well as generating new plans to cope with unforeseen situations at design time. AgentSpeak(PL) bridges the gap between agent languages and multiagent systems by introducing a simple cooperation mechanism together with a norm processing mechanism aimed at providing some degree of societal control. Publications in this area include:

MENEGUZZI, F. R.; ZORZO, Avelino Francisco; MÓRA, Michael da Costa; LUCK, Michael M.. Incorporating Planning into BDI Agents. Scalable Computing: Practice and Experience, v. 8, 2007.
[PDF] [BibTeX]
MENEGUZZI, F.; LUCK, Michael; Leveraging new plans in AgentSpeak(PL). Declarative Agent Languages and Technologies. Estoril, Portugal, 2008.
[PDF] [BibTeX] [Slides]
MENEGUZZI, F.; LUCK, Michael. Composing high-level plans for declarative agent programming. Declarative Agent Languages and Technologies. Honolulu, Hawai'i, 2007.
[PDF] [BibTeX] [Slides]
MENEGUZZI, F.R., ZORZO, Avelino Francisco, MÓRA, Michael da Costa. Mapping mental states into propositional planning. In: Proceedings of the 3rd International Joint Conference on Autonomous Agents and Multiagent Systems, New York, 2004.
[PDF] [BibTeX]
MENEGUZZI, F. R.; ZORZO, Avelino Francisco; MÓRA, Michael da Costa. Propositional Planning in BDI Agents. In: 19th Annual ACM Symposium on Applied Computing, 2004, Nicosia. Proceedings of the 19th Annual ACM Symposium on Applied Computing, 2004.
[PDF] [BibTeX]

AI Planning Formalisms

Planning is perhaps the first thing that comes to mind when one thinks about Artificial Intelligence. Understandably, algorithms for deciding what to do have been one of the most prolific areas of research in the field. One of the main contributions in my thesis was the integration of a planning algorithm to a BDI agent interpreter to strike a balance between human-optimized plans in a plan library and the creation of new plans at runtime to allow an agent to deal with situations not foreseen at design time. My interests lie in the relations that planning formalisms and algorithms have to agent reasoning, as well as the convertibility of deterministic planning formalisms (which are more straightforward for human planners) to stochastic planning formalisms. My work in the subject, in collaboration with Yuqing Tang and Simon Parsons has yielded the following publication (with others having been sent to various conferences at the moment):

MENEGUZZI, Felipe; TANG, Yuqing; SYCARA, Katia and PARSONS, Simon. An approach to generate MDPs using HTN representations, in Decision Making in Partially Observable, Uncertain Worlds: Exploring Insights from Multiple Communities (DMPOUW), Barcelona, Spain, 2011.
[PDF] [BibTeX] [Slides]
TANG, Yuqing; MENEGUZZI, Felipe; PARSONS, Simon and SYCARA, Katia. Probabilistic Hierarchical Planning over MDPs, in Proceedings of the 10th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Taipei, Taiwan, 2011.
[PDF] [BibTeX]
MENEGUZZI, Felipe; TANG, Yuqing; SYCARA, Katia and PARSONS, Simon. On representing planning domains under uncertainty, in The Fourth Annual Conference of the International Technology Alliance, London, UK, 2010.
[PDF] [BibTeX]

Proactive Assistance

Human users generally operate in complex dynamic environments where they face challenges due to cognitive overload in planning and replanning; that is, the users must perform multiple concurrent tasks including: collecting coherent information about the current situation, reasoning about constraints and policies, and dealing with uncertainty to achieve timely decision making. In order to help the users to cope with cognitive overload in such an environment, proactive agents can offer context-sensitive assistance by anticipating the users needs, autonomously planning assistive actions, and offering assistance in an appropriate format at a right time. While at CMU, I have worked on various architectures for proactive assistance in the context of information gathering for coalition operations. I also co-chaired (with Jean Oh) the AAAI Fall Symposium on Proactive Assistant Agents This work has yielded the following publications:

MENEGUZZI, Felipe; OH, Jean; CHAKRABORTY, Nilanjan; SYCARA, Katia; MEHROTRA, Siddharth; TITTLE, James and LEWIS, Michael. A cognitive architecture for emergency response, in Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Valencia, Spain, 2012.
[PDF] [BibTeX]
OH, Jean; MENEGUZZI, Felipe; SYCARA, Katia and NORMAN, Timothy J. An agent architecture for prognostic reasoning assistance, in Proceedings of the 22nd International Joint Conference on Artificial Intelligence (IJCAI), Barcelona, Spain, 2011.
[PDF] [BibTeX]
OH, Jean; MENEGUZZI, Felipe; SYCARA, Katia. Probabilistic plan recognition for intelligent information agents: Towards proactive software assistant agents, in Proceedings of the Third International Conference on Agents and Artificial Intelligence (ICAART), Rome, Italy, 2011.
[PDF] [BibTeX]
OH, Jean; MENEGUZZI, Felipe; SYCARA, Katia and NORMAN, Timothy J. Prognostic normative reasoning in coalition planning, in Proceedings of the 10th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Taipei, Taiwan, 2011.
[PDF] [BibTeX]
MENEGUZZI, Felipe; OH, Jean; SYCARA, Katia; PARSONS, Simon and NORMAN, Timothy J. Challenges in anticipatory information management under network constraints , in The Fourth Annual Conference of the International Technology Alliance, London, UK, 2010.
[PDF] [BibTeX]
MENEGUZZI, Felipe; OH, Jean and SYCARA, Katia. Intelligent information assistance for coalition operations , in Knowledge Systems for Coalition Operations, Vancouver, Canada, 2010.
[PDF] [BibTeX]
OH, Jean; MENEGUZZI, Felipe; SYCARA, Katia. ANTIPA: an agent architecture for intelligent information assistance, in Proceedings of the Nineteenth European Conference on Artificial Intelligence (ECAI), Lisbon, Portugal, 2010.
[PDF] [BibTeX]
OH, Jean; MENEGUZZI, Felipe; SYCARA, Katia and NORMAN, Timothy J. An agent architecture for intelligent information assistance, in The Fourth Annual Conference of the International Technology Alliance, London, UK, 2010.
[PDF] [BibTeX]