Felipe Meneguzzi

TAAS Journal 2026 Journal Article

A BDI Task-oriented Agent in Belief Space

• Alexandre Yukio Ichida
• Felipe Meneguzzi

Building conversational agents to help humans in domain-specific tasks is challenging since the agent needs to understand the natural language and act over it while accessing domain expert knowledge. Modern natural language processing techniques led to an expansion of conversational agents, with recent pretrained language models achieving increasingly accurate language recognition results using ever-larger open datasets. However, the black-box nature of such pretrained language models obscures the agent’s reasoning and its motivations when responding, leading to unexplained dialogues. In this work, we develop a Belief-desire-intention (BDI) agent as a task-oriented dialogue system to introduce mental attitudes similar to humans describing their behavior during a dialogue. We compare the BDI model with pipeline task-oriented dialogue system architecture by leveraging existing components from dialogue systems and developing the agent’s intention selection as a dialogue policy. We show that combining traditional agent modeling approaches, such as BDI, with more recent learning techniques can result in efficient and scrutable dialogue systems.

AAAI Conference 2026 Conference Paper

Hypertension and Total-Order Forward Decomposition Optimizations (Abstract Reprint)

• Maurício Cecílio Magnaguagno
• Felipe Meneguzzi
• Lavindra de Silva

Hierarchical Task Network (HTN) planners generate plans using a decomposition process with extra domain knowledge to guide search towards a planning task. Domain experts develop such domain knowledge through recipes of how to decompose higher level tasks, specifying which tasks can be decomposed and under what conditions. In most realistic domains, such recipes contain recursions, i.e., tasks that can be decomposed into other tasks that contain the original task. Such domains require that either the domain expert tailor such domain knowledge to the specific HTN planning algorithm, or an algorithm that can search efficiently using such domain knowledge. By leveraging a three-stage compiler design we can easily support more language descriptions and preprocessing optimizations that when chained can greatly improve runtime efficiency in such domains. In this paper we evaluate such optimizations with the HyperTensioN HTN planner, winner of the HTN IPC 2020 total-order track.

AAMAS Conference 2025 Conference Paper

Generalised BDI Planning

• Felipe Meneguzzi
• Ramon Fraga Pereira
• Nir Oren

Agent interpreters based on the Beliefs, Desires, and Intentions (BDI) model traditionally perform means-ends reasoning using plan libraries composed of reactive planning rules. However, the design of such rules often imposes a heavy knowledge engineering burden on a designer, and trades off flexibility for runtime efficiency. This use of planning rules originates from the limitations of planning technology at the time of the first BDI implementations. While these limitations have gradually been overcome by the integration of various types of planning into existing BDI theories, the corresponding interpreters remain fundamentally plan-library based. In this paper, we develop a novel BDI agent architecture driven by generalised planning as means-ends reasoning, in a radical departure from existing architectures. This architecture has two key properties. First, it more closely resembles the foundations of BDI logic and reasoning. Second, it offers substantial gains in efficiency in comparison with an architecture driven by classical planning.

JAAMAS Journal 2025 Journal Article

Hypertension and total-order forward decomposition optimizations

• Maurício Cecílio Magnaguagno
• Felipe Meneguzzi
• Lavindra de Silva

Abstract Hierarchical Task Network (HTN) planners generate plans using a decomposition process with extra domain knowledge to guide search towards a planning task. Domain experts develop such domain knowledge through recipes of how to decompose higher level tasks, specifying which tasks can be decomposed and under what conditions. In most realistic domains, such recipes contain recursions, i. e. , tasks that can be decomposed into other tasks that contain the original task. Such domains require that either the domain expert tailor such domain knowledge to the specific HTN planning algorithm, or an algorithm that can search efficiently using such domain knowledge. By leveraging a three-stage compiler design we can easily support more language descriptions and preprocessing optimizations that when chained can greatly improve runtime efficiency in such domains. In this paper we evaluate such optimizations with the HyperTensioN HTN planner, winner of the HTN IPC 2020 total-order track.

AAMAS Conference 2025 Conference Paper

Intention Recognition in Real-Time Interactive Navigation Maps

• Peijie Zhao
• Zunayed Arefin
• Felipe Meneguzzi
• Ramon Fraga Pereira

In this demonstration, we develop IntentRec4Maps, a system to recognise users’ intentions in real-time interactive navigation maps. IntentRec4Maps uses the Google Maps Platform as the realworld interactive map, and a well-known approach for recognising intentions in real-time. We showcase IntentRec4Maps using two different Path-Planners and a Large Language Model (LLM).

ECAI Conference 2024 Conference Paper

A Practical Operational Semantics for Classical Planning in BDI Agents

• Mengwei Xu 0002
• Tom Lumley
• Ramon Fraga Pereira
• Felipe Meneguzzi

Implementations of the Belief-Desire-Intention (BDI) architecture have a long tradition in the development of autonomous agent systems. However, most practical implementations of the BDI framework rely on a pre-defined plan library for decision-making, which places a significant burden on programmers, and still yields systems that may be brittle, struggling to achieve their goals in dynamic environments. This paper overcomes this limitation by introducing an operational semantics for BDI systems that rely on Classical Planning at run time to both cope with failures that were unforeseeable and synthesise new plans that were unspecified at design time. This semantics places particular emphasis on the interaction of the reasoning cycle and an underlying planning algorithm. We empirically demonstrate the practical feasibility and generality of such an approach in an implementation of this semantics within two popular BDI platforms together with in-depth computational evaluation.

IJCAI Conference 2024 Conference Paper

A Survey on Model-Free Goal Recognition

• Leonardo Amado
• Sveta Paster Shainkopf
• Ramon Fraga Pereira
• Reuth Mirsky
• Felipe Meneguzzi

Goal Recognition is the task of inferring an agent's intentions from a set of observations. Existing recognition approaches have made considerable advances in domains such as human-robot interaction, intelligent tutoring systems, and surveillance. However, most approaches rely on explicit domain knowledge, often defined by a domain expert. Much recent research focus on mitigating the need for a domain expert while maintaining the ability to perform quality recognition, leading researchers to explore Model-Free Goal Recognition approaches. We comprehensively survey Model-Free Goal Recognition, and provide a perspective on the state-of-the-art approaches and their applications, showing recent advances. We categorize different approaches, introducing a taxonomy with a focus on their characteristics, strengths, weaknesses, and suitability for different scenarios. We compare the advances each approach made to the state-of-the-art and provide a direction for future research in Model-Free Goal Recognition.

AAMAS Conference 2024 Conference Paper

BDI Agents in Natural Language Environments

• Alexandre Yukio Ichida
• Felipe Meneguzzi
• Rafael C. Cardoso

Developing autonomous agents to deal with real-world problems is challenging, especially when developers are not necessarily specialists in artificial intelligence. This poses two key challenges regarding the interface of the programming with the developer, and the efficiency of the resulting agents. In this paper we tackle both challenges in an efficient agent architecture that leverages recent developments in natural language processing, and the intuitive folk psychology abstraction of the beliefs, desires, intentions (BDI) architecture. The resulting architecture uses existing reinforcement learning techniques to bootstrap the agent’s reasoning capabilities while allowing a developer to instruct the agent more directly using natural language as its programming interface. We empirically show the efficiency gains of natural language plans over a pure machine learning approach in the ScienceWorld environment.

AAMAS Conference 2024 Conference Paper

Empowering BDI Agents with Generalised Decision-Making

• Ramon Fraga Pereira
• Felipe Meneguzzi

While research on software agents has long focused on explicit agent communication, there is comparatively less effort on implicit communication between agents via recognising each other’s intentions and desires for understanding their decision-making reasoning process. Since most human communication is not explicit, we aim to outline a research agenda to help endow autonomous agents with analogous coordination capabilities. In this paper, we formalise a framework that empowers the decision-making process of BDI agents in adversarial and cooperative environments by casting them as generalised planners using Theory of Mind. Our formalisation uses the fundamental philosophical properties of the BDI model and its reasoning process to outline a broad research agenda in agents’ research.

ECAI Conference 2024 Conference Paper

Explorative Imitation Learning: A Path Signature Approach for Continuous Environments

• Nathan Gavenski
• Juarez Monteiro
• Felipe Meneguzzi
• Michael Luck
• Odinaldo Rodrigues

Some imitation learning methods combine behavioural cloning with self-supervision to infer actions from state pairs. However, most rely on a large number of expert trajectories to increase generalisation and human intervention to capture key aspects of the problem, such as domain constraints. In this paper, we propose Continuous Imitation Learning from Observation (CILO), a new method augmenting imitation learning with two important features: (i) exploration, allowing for more diverse state transitions, requiring less expert trajectories and resulting in fewer training iterations; and (ii) path signatures, allowing for automatic encoding of constraints, through the creation of non-parametric representations of agents and expert trajectories. We compared CILO with a baseline and two leading imitation learning methods in five environments. It had the best overall performance of all methods in all environments, outperforming the expert in two of them.

ECAI Conference 2024 Conference Paper

Real-Time Goal Recognition Using Approximations in Euclidean Space

• Douglas Antunes Tesch
• Leonardo Amado
• Felipe Meneguzzi

While recent work on online goal recognition efficiently infers goals under low observability, comparatively less work focuses on online goal recognition that works in both discrete and continuous domains. Online goal recognition approaches often rely on repeated calls to the planner at each new observation, incurring high computational costs. Recognizing goals online in continuous space quickly and reliably is critical for any trajectory planning problem since the real physical world is fast-moving, e. g. robot applications. We develop an efficient method for goal recognition that relies either on a single call to the planner for each possible goal in discrete domains or a simplified motion model that reduces the computational burden in continuous ones. The resulting approach performs the online component of recognition orders of magnitude faster than the current state of the art, making it the first online method effectively usable for robotics applications that require sub-second recognition.

IJCAI Conference 2023 Conference Paper

Multi-Agent Intention Recognition and Progression

• Michael Dann
• Yuan Yao
• Natasha Alechina
• Brian Logan
• Felipe Meneguzzi
• John Thangarajah

For an agent in a multi-agent environment, it is often beneficial to be able to predict what other agents will do next when deciding how to act. Previous work in multi-agent intention scheduling assumes a priori knowledge of the current goals of other agents. In this paper, we present a new approach to multi-agent intention scheduling in which an agent uses online goal recognition to identify the goals currently being pursued by other agents while acting in pursuit of its own goals. We show how online goal recognition can be incorporated into an MCTS-based intention scheduler, and evaluate our approach in a range of scenarios. The results demonstrate that our approach can rapidly recognise the goals of other agents even when they are pursuing multiple goals concurrently, and has similar performance to agents which know the goals of other agents a priori.

AAAI Conference 2023 Conference Paper

Robust Neuro-Symbolic Goal and Plan Recognition

• Leonardo Amado
• Ramon Fraga Pereira
• Felipe Meneguzzi

Goal Recognition is the task of discerning the intended goal of an agent given a sequence of observations, whereas Plan Recognition consists of identifying the plan to achieve such intended goal. Regardless of the underlying techniques, most recognition approaches are directly affected by the quality of the available observations. In this paper, we develop neuro-symbolic recognition approaches that can combine learning and planning techniques, compensating for noise and missing observations using prior data. We evaluate our approaches in standard human-designed planning domains as well as domain models automatically learned from real-world data. Empirical experimentation shows that our approaches reliably infer goals and compute correct plans in the experimental datasets. An ablation study shows that outperform approaches that rely exclusively on the domain model, or exclusively on machine learning in problems with both noisy observations and low observability.

AAAI Conference 2022 Conference Paper

Goal Recognition as Reinforcement Learning

• Leonardo Amado
• Reuth Mirsky
• Felipe Meneguzzi

Most approaches for goal recognition rely on specifications of the possible dynamics of the actor in the environment when pursuing a goal. These specifications suffer from two key issues. First, encoding these dynamics requires careful design by a domain expert, which is often not robust to noise at recognition time. Second, existing approaches often need costly real-time computations to reason about the likelihood of each potential goal. In this paper, we develop a framework that combines model-free reinforcement learning and goal recognition to alleviate the need for careful, manual domain design, and the need for costly online executions. This framework consists of two main stages: offline learning of policies or utility functions for each potential goal, and online inference. We provide a first instance of this framework using tabular Q-learning for the learning stage, as well as three mechanisms for the inference stage. The resulting instantiation achieves state-of-the-art performance against goal recognizers on standard evaluation domains and superior performance in noisy environments.

IJCAI Conference 2021 Conference Paper

A Survey on Goal Recognition as Planning

• Felipe Meneguzzi
• Ramon Fraga Pereira

Goal Recognition is the task of inferring an agent's goal, from a set of hypotheses, given a model of the environment dynamic, and a sequence of observations of such agent's behavior. While research on this problem gathered momentum as an offshoot of plan recognition, recent research has established it as a major subject of research on its own, leading to numerous new approaches that both expand the expressivity of domains in which to perform goal recognition and substantial advances to the state-of-the-art on established domain types. In this survey, we focus on the advances to goal recognition achieved in the last decade, categorizing the resulting techniques and identifying a number of opportunities for further breakthrough research.

AAAI Conference 2021 Conference Paper

An LP-Based Approach for Goal Recognition as Planning

• Luísa R. A. Santos
• Felipe Meneguzzi
• Ramon Fraga Pereira
• André Grahl Pereira

Goal recognition aims to recognize the set of candidate goals that are compatible with the observed behavior of an agent. In this paper, we develop a method based on the operatorcounting framework that efficiently computes solutions that satisfy the observations and uses the information generated to solve goal recognition tasks. Our method reasons explicitly about both partial and noisy observations: estimating uncertainty for the former, and satisfying observations given the unreliability of the sensor for the latter. We evaluate our approach empirically over a large data set, analyzing its components on how each can impact the quality of the solutions. In general, our approach is superior to previous methods in terms of agreement ratio, accuracy, and spread. Finally, our approach paves the way for new research on combinatorial optimization to solve goal recognition tasks.

ICAPS Conference 2021 Conference Paper

Automated Design of fMRI Paradigms

• Katherine Bianchini Esper
• Felipe Meneguzzi

Neuroimaging techniques have been widely used in recent decades to assess brain activation patterns for neuroscience. Task design is the most important challenge for neuroimaging studies, to achieve the best modeling for assessing brain patterns within and across subjects. Specifically, functional magnetic resonance imaging (fMRI) experiments rely on the precise and effective design of sequences of stimuli intended to activate specific brain regions (i. e. paradigm design). In this paper, we use PDDL+ to model fMRI paradigms so that neuroscientists can use automated planning to design neuroimaging paradigms in a declarative way. Planning neuroimaging paradigms is especially important for functional studies and presurgical planning. The former should help to ensure an experimental design that allows the analysis of the brain regions that are interesting in the study. The latter should help surgeons select the correct stimuli for a presurgical, non-invasive, exploration of the cognitive functions that might be affected by debridement of brain lesions.

AAMAS Conference 2021 Conference Paper

Combining LSTMs and Symbolic Approaches for Robust Plan Recognition

• Leonardo Rosa Amado
• Ramon Fraga Pereira
• Felipe Meneguzzi

IJCAI Conference 2020 Conference Paper

BDI Agent Architectures: A Survey

• Lavindra de Silva
• Felipe Meneguzzi
• Brian Logan

The BDI model forms the basis of much of the research on symbolic models of agency and agent-oriented software engineering. While many variants of the basic BDI model have been proposed in the literature, there has been no systematic review of research on BDI agent architectures in over 10 years. In this paper, we survey the main approaches to each component of the BDI architecture, how these have been realised in agent programming languages, and discuss the trade-offs inherent in each approach.

AIJ Journal 2020 Journal Article

Landmark-based approaches for goal recognition as planning

• Ramon Fraga Pereira
• Nir Oren
• Felipe Meneguzzi

AAAI Conference 2020 Short Paper

LatRec: Recognizing Goals in Latent Space (Student Abstract)

• Leonardo Amado
• Felipe Meneguzzi

Recent approaches to goal recognition have progressively relaxed the requirements about the amount of domain knowledge and available observations, yielding accurate and ef- ficient algorithms. These approaches, however, assume that there is a domain expert capable of building complete and correct domain knowledge to successfully recognize an agent’s goal. This is too strong for most real-world applications. We overcome these limitations by combining goal recognition techniques from automated planning, and deep autoencoders to carry out unsupervised learning to generate domain theories from data streams and use the resulting domain theories to deal with incomplete and noisy observations. Moving forward, we aim to develop a new data-driven goal recognition technique that infers the domain model using the same set of observations used in recognition itself.

AAAI Conference 2020 Conference Paper

Semantic Attachments for HTN Planning

• Maurício Cecílio Magnaguagno
• Felipe Meneguzzi

Hierarchical Task Networks (HTN) planning uses a decomposition process guided by domain knowledge to guide search towards a planning task. While many HTN planners allow calls to external processes (e. g. to a simulator interface) during the decomposition process, this is a computationally expensive process, so planner implementations often use such calls in an ad-hoc way using very specialized domain knowledge to limit the number of calls. Conversely, the classical planners that are capable of using external calls (often called semantic attachments) during planning are limited to generating a fixed number of ground operators at problem grounding time. We formalize Semantic Attachments for HTN planning using semi coroutines, allowing such procedurally de- fined predicates to link the planning process to custom unifications outside of the planner, such as numerical results from a robotics simulator. The resulting planner then uses such coroutines as part of its backtracking mechanism to search through parallel dimensions of the state-space (e. g. through numeric variables). We show empirically that our planner outperforms the state-of-the-art numeric planners in a number of domains using minimal extra domain knowledge.

TIST Journal 2020 Journal Article

Using Sub-Optimal Plan Detection to Identify Commitment Abandonment in Discrete Environments

• Ramon Fraga Pereira
• Nir Oren
• Felipe Meneguzzi

Assessing whether an agent has abandoned a goal or is actively pursuing it is important when multiple agents are trying to achieve joint goals, or when agents commit to achieving goals for each other. Making such a determination for a single goal by observing only plan traces is not trivial, as agents often deviate from optimal plans for various reasons, including the pursuit of multiple goals or the inability to act optimally. In this article, we develop an approach based on domain independent heuristics from automated planning, landmarks, and fact partitions to identify sub-optimal action steps—with respect to a plan—within a fully observable plan execution trace. Such capability is very important in domains where multiple agents cooperate and delegate tasks among themselves, such as through social commitments, and need to ensure that a delegating agent can infer whether or not another agent is actually progressing towards a delegated task. We demonstrate how a creditor can use our technique to determine—by observing a trace—whether a debtor is honouring a commitment. We empirically show, for a number of representative domains, that our approach infers sub-optimal action steps with very high accuracy and detects commitment abandonment in nearly all cases.

AAMAS Conference 2019 Conference Paper

Classification of Contractual Conflicts via Learning of Semantic Representations

• João Paulo Aires
• Roger Granada
• Juarez Monteiro
• Rodrigo Coelho Barros
• Felipe Meneguzzi

Contracts are the main medium through which parties formalize their trade relations, be they the exchange of goods or the specification of mutual obligations. While electronic contracts allow automated processes to verify their correctness, most agreements in the real world are still written in natural language, which need substantial human revision effort to eliminate possible conflicting statements in long and complex contracts. In this paper, we formalize a typology of conflict types between clauses suitable for machine learning and develop techniques to review contracts by learning to identify and classify such conflicts, facilitating the task of contract revision. We evaluate the effectiveness of our techniques using a manually annotated contract conflict corpus with results close to the current state-of-the-art for conflict identification, while introducing a more complex classification task of such conflicts for which our method surpasses the state-of-the art method.

AAMAS Conference 2019 Conference Paper

ConCon: A Contract Conflict Identifier

• João Paulo Aires
• Roger Granada
• Felipe Meneguzzi

Contracts are the main medium through which people and legal entities formalise their trade relations, be they the exchange of goods or the specification of mutual obligations. While electronic contracts allow automated processes to verify their correctness, most agreements in the real world are still encoded in contracts written in natural language, necessitating substantial human revision effort to eliminate possible conflicting statements, especially for long and complex contracts. We demonstrate the ConCon (Contract Conflicts) tool, to automatically read natural language contracts and indicate potential conflicts among their clauses. Using our tool, legal professionals and the general public can benefit from a ranking of potential conflicts between the clauses in a contract, saving time and effort from legal experts in contract proof-reading.

ICAPS Conference 2019 Conference Paper

Landmark-Enhanced Heuristics for Goal Recognition in Incomplete Domain Models

• Ramon Fraga Pereira
• André Grahl Pereira
• Felipe Meneguzzi

Recent approaches to goal recognition have progressively relaxed the assumptions about the amount and correctness of domain knowledge and available observations, yielding accurate and efficient algorithms. These approaches, however, assume completeness and correctness of the domain theory against which their algorithms match observations: this is too strong for most real-world domains. In this paper, we develop goal recognition techniques that are capable of recognizing goals using incomplete domain theories by considering different notions of planning landmarks in such domains. We evaluate the resulting techniques empirically in a large dataset of incomplete domains, and perform an ablation study to understand their effect on recognition performance.

IJCAI Conference 2019 Conference Paper

Online Probabilistic Goal Recognition over Nominal Models

• Ramon Fraga Pereira
• Mor Vered
• Felipe Meneguzzi
• Miquel Ramírez

This paper revisits probabilistic, model-based goal recognition to study the implications of the use of nominal models to estimate the posterior probability distribution over a finite set of hypothetical goals. Existing model-based approaches rely on expert knowledge to produce symbolic descriptions of the dynamic constraints domain objects are subject to, and these are assumed to produce correct predictions. We abandon this assumption to consider the use of nominal models that are learnt from observations on transitions of systems with unknown dynamics. Leveraging existing work on the acquisition of domain models via learning for Hybrid Planning we adapt and evaluate existing goal recognition approaches to analyze how prediction errors, inherent to system dynamics identification and model learning techniques have an impact over recognition error rates.

IJCAI Conference 2018 Conference Paper

An Operational Semantics for a Fragment of PRS

• Lavindra de Silva
• Felipe Meneguzzi
• Brian Logan

The Procedural Reasoning System (PRS) is arguably the first implementation of the Belief--Desire--Intention (BDI) approach to agent programming. PRS remains extremely influential, directly or indirectly inspiring the development of subsequent BDI agent programming languages. However, perhaps surprisingly given its centrality in the BDI paradigm, PRS lacks a formal operational semantics, making it difficult to determine its expressive power relative to other agent programming languages. This paper takes a first step towards closing this gap, by giving a formal semantics for a significant fragment of PRS. We prove key properties of the semantics relating to PRS-specific programming constructs, and show that even the fragment of PRS we consider is strictly more expressive than the plan constructs found in typical BDI languages.

AAAI Conference 2018 Short Paper

Goal Recognition in Incomplete Domain Models

• Ramon Pereira
• Felipe Meneguzzi

Recent approaches to goal recognition have progressively relaxed the assumptions about the amount and correctness of domain knowledge and available observations, yielding accurate and efficient algorithms. These approaches, however, assume completeness and correctness of the domain theory against which their algorithms match observations: this is too strong for most real-world domains. In this work, we develop a goal recognition technique capable of recognizing goals using incomplete (and possibly incorrect) domain theories.

JAAMAS Journal 2018 Journal Article

GoCo: planning expressive commitment protocols

• Felipe Meneguzzi
• Mauricio C. Magnaguagno
• Neil Yorke-Smith

Abstract This article addresses the challenge of planning coordinated activities for a set of autonomous agents, who coordinate according to social commitments among themselves. We develop a multi-agent plan in the form of a commitment protocol that allows the agents to coordinate in a flexible manner, retaining their autonomy in terms of the goals they adopt so long as their actions adhere to the commitments they have made. We consider an expressive first-order setting with probabilistic uncertainty over action outcomes. We contribute the first practical means to derive protocol enactments which maximise expected utility from the point of view of one agent. Our work makes two main contributions. First, we show how Hierarchical Task Network planning can be used to enact a previous semantics for commitment and goal alignment, and we extend that semantics in order to enact first-order commitment protocols. Second, supposing a cooperative setting, we introduce uncertainty in order to capture the reality that an agent does not know for certain that its partners will successfully act on their part of the commitment protocol. Altogether, we employ hierarchical planning techniques to check whether a commitment protocol can be enacted efficiently, and generate protocol enactments under a variety of conditions. The resulting protocol enactments can be optimised either for the expected reward or the probability of a successful execution of the protocol. We illustrate our approach on a real-world healthcare scenario.

YNICL Journal 2018 Journal Article

Identification of autism spectrum disorder using deep learning and the ABIDE dataset

• Anibal Sólon Heinsfeld
• Alexandre Rosa Franco
• R. Cameron Craddock
• Augusto Buchweitz
• Felipe Meneguzzi

KER Journal 2018 Journal Article

Q-Table compression for reinforcement learning

• Leonardo Amado
• Felipe Meneguzzi

Abstract Reinforcement learning (RL) algorithms are often used to compute agents capable of acting in environments without prior knowledge of the environment dynamics. However, these algorithms struggle to converge in environments with large branching factors and their large resulting state-spaces. In this work, we develop an approach to compress the number of entries in a Q-value table using a deep auto-encoder. We develop a set of techniques to mitigate the large branching factor problem. We present the application of such techniques in the scenario of a real-time strategy (RTS) game, where both state space and branching factor are a problem. We empirically evaluate an implementation of the technique to control agents in an RTS game scenario where classical RL fails and provide a number of possible avenues of further work on this problem.

AAMAS Conference 2018 Conference Paper

Towards Online Goal Recognition Combining Goal Mirroring and Landmarks

• Mor Vered
• Ramon Fraga Pereira
• Maur�cio C. Magnaguagno
• Gal A. Kaminka
• Felipe Meneguzzi

AAMAS Conference 2017 Conference Paper

A Deep Learning Approach for Norm Conflict Identification

• Joã o Paulo Aires
• Felipe Meneguzzi

AAMAS Conference 2017 Conference Paper

Detecting Commitment Abandonment by Monitoring Sub-Optimal Steps during Plan Execution

• Ramon Fraga Pereira
• Nir Oren
• Felipe Meneguzzi

AAAI Conference 2017 Conference Paper

Landmark-Based Heuristics for Goal Recognition

• Ramon Pereira
• Nir Oren
• Felipe Meneguzzi

Automated planning can be used to efficiently recognize goals and plans from partial or full observed action sequences. In this paper, we propose goal recognition heuristics that rely on information from planning landmarks — facts or actions that must occur if a plan is to achieve a goal when starting from some initial state. We develop two such heuristics: the first estimates goal completion by considering the ratio between achieved and extracted landmarks of a candidate goal, while the second takes into account how unique each landmark is among landmarks for all candidate goals. We empirically evaluate these heuristics over both standard goal/plan recognition problems, and a set of very large problems. We show that our heuristics can recognize goals more accurately, and run orders of magnitude faster, than the current state-of-the-art.

AILAW Journal 2017 Journal Article

Norm conflict identification in contracts

• João Paulo Aires
• Daniele Pinheiro
• Vera Strube de Lima
• Felipe Meneguzzi

Abstract The exchange of goods and services between individuals is often formalised by a contract in which the parties establish norms to define what is expected of each one. Norms use deontic statements of obligation, prohibition, and permission, which may be in conflict. The task of manually detecting norm conflicts can be time–consuming and error-prone since contracts can be vast and complex. To automate such tasks, we develop an approach to identify potential conflicts between norms. We show the effectiveness of our approach and its individual components empirically using two publicly available corpora, and contribute with a new annotated test corpus for norm conflict identification.

ECAI Conference 2016 Conference Paper

A Bayesian Approach to Norm Identification

• Stephen Cranefield
• Felipe Meneguzzi
• Nir Oren
• Bastin Tony Roy Savarimuthu

When entering a system, an agent should be aware of the obligations and prohibitions (collectively norms) that affect it. Existing solutions to this norm identification problem make use of observations of either norm compliant, or norm violating, behaviour. Thus, they assume an extreme situation where norms are typically violated, or complied with. In this paper we propose a Bayesian approach to norm identification which operates by learning from both norm compliant and norm violating behaviour. We evaluate our approach's effectiveness empirically and compare its accuracy to existing approaches. By utilising both types of behaviour, we not only overcome a major limitation of such approaches, but also obtain improved performance over the state of the art, allowing norms to be learned with fewer observations.

ECAI Conference 2016 Conference Paper

Landmark-Based Plan Recognition

• Ramon Fraga Pereira
• Felipe Meneguzzi

EAAI Journal 2015 Journal Article

BDI reasoning with normative considerations

• Felipe Meneguzzi
• Odinaldo Rodrigues
• Nir Oren
• Wamberto W. Vasconcelos
• Michael Luck

AILAW Journal 2015 Journal Article

Monitoring compliance with E-contracts and norms

• Sanjay Modgil
• Nir Oren
• Noura Faci
• Felipe Meneguzzi
• Simon Miles
• Michael Luck

Abstract The behaviour of autonomous agents may deviate from that deemed to be for the good of the societal systems of which they are a part. Norms have therefore been proposed as a means to regulate agent behaviours in open and dynamic systems, where these norms specify the obliged, permitted and prohibited behaviours of agents. Regulation can effectively be achieved through use of enforcement mechanisms that result in a net loss of utility for an agent in cases where the agent’s behaviour fails to comply with the norms. Recognition of compliance is thus crucial for achieving regulation. In this paper, we propose a general framework for observation of agents’ behaviour, and recognition of this behaviour as constituting, or counting as, compliance or violation. The framework deploys monitors that receive inputs from trusted observers, and processes these inputs together with transition network representations of individual norms. In this way, monitors determine the fulfillment or violation status of norms. The paper also describes a proof of concept implementation of the framework, and its deployment in electronic contracting environments.

ECAI Conference 2014 Conference Paper

Analyzing the tradeoff between efficiency and cost of norm enforcement in stochastic environments

• Moser Silva Fagundes
• Sascha Ossowski
• Felipe Meneguzzi

In multiagent systems, agents might interfere with each other as a side-effect of their activities. One approach to coordinating these agents is to restrict their activities by means of social norms whose violation results in sanctions to violating agents. We formalize a normative system within a stochastic environment and norm enforcement follows a stochastic model in which stricter enforcement entails higher cost. Within this type of system, we provide an approach to analize the tradeoff between norm enforcement efficiency and its cost considering a population of norm-aware selfish agents.

AAAI Conference 2013 Conference Paper

A First-Order Formalization of Commitments and Goals for Planning

• Felipe Meneguzzi
• Pankaj Telang
• Munindar Singh

Commitments help model interactions in multiagent systems in a computationally realizable yet high-level manner without compromising the autonomy and heterogeneity of the member agents. Recent work shows how to combine commitments with goals and apply planning methods to enable agents to determine their actions. However, previous approaches to modeling commitments are confined to propositional representations, which limits their applicability in practical cases. We propose a first-order representation and reasoning technique that accommodates templatic commitments and goals that may be applied repeatedly with differing bindings for domain objects. Doing so not only leads to a more perspicuous modeling, but also supports many practical patterns.

KER Journal 2013 Journal Article

Planning in BDI agents: a survey of the integration of planning algorithms and agent reasoning

• Felipe Meneguzzi
• Lavindra de Silva

Abstract Agent programming languages have often avoided the use of automated (first principles or hierarchical) planners in favour of predefined plan/recipe libraries for computational efficiency reasons. This allows for very efficient agent reasoning cycles, but limits the autonomy and flexibility of the resulting agents, oftentimes with deleterious effects on the agent's performance. Planning agents can, for instance, synthesise a new plan to achieve a goal for which no predefined recipe worked, or plan to make viable the precondition of a recipe belonging to a goal being pursued. Recent work on integrating automated planning with belief-desire-intention (BDI)-style agent architectures has yielded a number of systems and programming languages that exploit the efficiency of standard BDI reasoning, as well as the flexibility of generating new recipes at runtime. In this paper, we survey these efforts and point out directions for future work.

EAAI Journal 2013 Journal Article

Prognostic normative reasoning

• Jean Oh
• Felipe Meneguzzi
• Katia Sycara
• Timothy J. Norman

AAMAS Conference 2012 Conference Paper

A cognitive architecture for emergency response

• Felipe Meneguzzi
• Siddharth Mehrotra
• James Tittle
• Jean Oh
• Nilanjan Chakraborty
• Katia Sycara
• Michael Lewis

Plan recognition, cognitive workload estimation and human assistance have been extensively studied in the AI and human factors communities, but have seldom been integrated and evaluated as complete systems. In this paper, we develop an assistant agent architecture integrating plan recognition, current and future user information needs, workload estimation and adaptive information presentation to aid an emergency response manager in making high quality decisions under time stress, while avoiding cognitive overload. We describe its main components as well as results for en experiment simulating various possible executions of the emergency response plans used in the real world, comparing reaction time of an assisted versus an unassisted human.

EAAI Journal 2012 Journal Article

Applying electronic contracting to the aerospace aftercare domain

• Felipe Meneguzzi
• Sanjay Modgil
• Nir Oren
• Simon Miles
• Michael Luck
• Noura Faci

AAAI Conference 2012 Conference Paper

Querying Linked Ontological Data through Distributed Summarization

• Achille Fokoue
• Felipe Meneguzzi
• Murat Sensoy
• Jeff Pan

As the semantic web expands, ontological data becomes distributed over a large network of data sources on the Web. Consequently, evaluating queries that aim to tap into this distributed semantic database necessitates the ability to consult multiple data sources efficiently. In this paper, we propose methods and heuristics to efficiently query distributed ontological data based on a series of properties of summarized data. In our approach, each source summarizes its data as another RDF graph, and relevant section of these summaries are merged and analyzed at query evaluation time. We show how the analysis of these summaries enables more efficient source selection, query pruning and transformation of expensive distributed joins into local joins.

IJCAI Conference 2011 Conference Paper

An Agent Architecture for Prognostic Reasoning Assistance

• Jean Oh
• Felipe Meneguzzi
• Katia Sycara
• Timothy J. Norman

In this paper we describe a software assistant agent that can proactively assist human users situated in a time-constrained environment to perform normative reasoning--reasoning about prohibitions and obligations--so that the user can focus on her planning objectives. In order to provide proactive assistance, the agent must be able to 1) recognize the user's planned activities, 2) reason about potential needs of assistance associated with those predicted activities, and 3) plan to provide appropriate assistance suitable for newly identified user needs. To address these specific requirements, we develop an agent architecture that integrates user intention recognition, normative reasoning over a user's intention, and planning, execution and replanning for assistive actions. This paper presents the agent architecture and discusses practical applications of this approach.

AAMAS Conference 2011 Conference Paper

Probabilistic Hierarchical Planning over MDPs

• Yuqing Tang
• Felipe Meneguzzi
• Katia Sycara
• Simon Parsons

In this paper, we propose a new approach to using probabilistic hierarchical task networks (HTNs) as an effective method for agents to plan in conditions in which their problem-solving knowledge is uncertain, and the environment is non-deterministic. In such situations it is natural to model the environment as a Markov decision process (MDP). We show that using Earley graphs, it is possible to bridge the gap between HTNs and MDPs. We prove that the size of the Earley graph created for given HTNs is bounded by the total number of tasks in the HTNs and show that from the Earley graph we can then construct a plan for a given task that has the maximum expected value when it is executed in an MDP environment.

AAMAS Conference 2011 Conference Paper

Prognostic Normative Reasoning in Coalition Planning

• Jean Oh
• Felipe Meneguzzi
• Katia Sycara
• Timothy J. Norman

In this paper we describe a software assistant agent that can proactively assist human users situated in a time-constrained coalition environment. The cognitive workload is significantly increased when the user must not only cope with a complex environment, but also with a set of unaccustomed rules that prescribe how the coalition planning process must be carried out. In this context, we introduce the notion of prognostic norm reasoning to predict the user's likely normative violations, allowing the assistant agent to plan and take remedial actions before the violations actually occur. To the best of our knowledge, our approach is the first that manages norms in a proactive and autonomous manner.

ECAI Conference 2010 Conference Paper

ANTIPA: an agent architecture for intelligent information assistance

• Jean Oh
• Felipe Meneguzzi
• Katia P. Sycara

Human users trying to plan and accomplish information-dependent goals in highly dynamic environments with prevalent uncertainty must consult various types of information sources in their decision-making processes while the information requirements change as they plan and re-plan. When the users must make time-critical decisions in information-intensive tasks they become cognitively overloaded not only by the planning activities but also by the information-gathering activities at various points in the planning process. We have developed the ANTicipatory Information and Planning Agent (ANTIPA) to manage information adaptively in order to mitigate user cognitive overload. To this end, the agent brings information to the user as a result of user requests but most crucially, it proactively predicts the user's prospective information needs by recognizing the user's plan; pre-fetches information that is likely to be used in the future; and offers the information when it is relevant to the current or future planning decisions. This paper introduces a fully implemented agent of the ANTIPA architecture using a decision-theoretic user model.

AAMAS Conference 2009 Conference Paper

A Framework for Monitoring Agent-Based Normative Systems

• Sanjay Modgil
• Noura Faci
• Felipe Meneguzzi
• Nir Oren
• Simon Miles
• Michael Luck

The behaviours of autonomous agents may deviate from those deemed to be for the good of the societal systems of which they are a part. Norms have therefore been proposed as a means to regulate agent behaviours in open and dynamic systems, where these norms specify the obliged, permitted and prohibited behaviours of agents. Regulation can effectively be achieved through use of enforcement mechanisms that result in a net loss of utility for an agent in cases where the agent’s behaviour fails to comply with the norms. Recognition of compliance is thus crucial for achieving regulation. In this paper we propose a generic architecture for observation of agent behaviours, and recognition of these behaviours as constituting, or counting as, compliance or violation. The architecture deploys monitors that receive inputs from observers, and processes these inputs together with transition network representations of individual norms. In this way, monitors determine the fulfillment or violation status of norms. The paper also describes a proof of concept implementation and deployment of monitors in electronic contracting environments.

AAMAS Conference 2009 Conference Paper

Norm-Based Behaviour Modification in Bdi Agents

• Felipe Meneguzzi
• Michael Luck

While there has been much work on developing frameworks and models of norms and normative systems, consideration of the impact of norms on the practical reasoning of agents has attracted less attention. The problem is that traditional agent architectures and their associated languages provide no mechanism to adapt an agent at runtime to norms constraining their behaviour. This is important because if BDI-type agents are to operate in open environments, they need to adapt to changes in the norms that regulate such environments. In response, in this paper we provide a technique to extend BDI agent languages, by enabling them to enact behaviour modification at runtime in response to newly accepted norms. Our solution consists of creating new plans to comply with obligations and suppressing the execution of existing plans that violate prohibitions. We demonstrate the viability of our approach through an implementation of our solution in the AgentSpeak(L) language.

AAMAS Conference 2008 Conference Paper

Electronic contracting in aircraft aftercare: A case study

• Felipe Meneguzzi
• Simon Miles
• Michael Luck
• Camden Holt
• Malcolm Smith

Distributed systems comprised of autonomous self-interested entities require some sort of control mechanism to ensure the predictability of the interactions that drive them. This is certainly true in the aerospace domain, where manufacturers, suppliers and operators must coordinate their activities to maximise safety and profit, for example. To address this need, the notion of norms has been proposed which, when incorporated into formal electronic documents, allow for the specification and deployment of contractdriven systems. In this context, we describe the CONTRACT framework and architecture for exactly this purpose, and describe a concrete instantiation of this architecture as a prototype system applied to an aerospace aftercare scenario.