Scaling up CCG-Based Plan Recognition via Monte-Carlo Tree Search

Kantharaju, Pavan; Ontañón, Santiago; Geib, Christopher W.

doi:10.1109/cig.2019.8848013

Cited by 7 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Experiments for behaviour classification were conducted on a dataset comprising replays of different types of scripted agents in the real-time strategy game StarCraft (Kantharaju, Ontañón, and Geib 2019), and on two real-world malware datasets comprising 'actions' taken by different malware applications in response to various Android system events (BootCompleted and BatteryLow) (Bernardi et al 2019). The behaviour classification task involves predicting the type of StarCraft agent and malware family, respectively, that generated a given behaviour trace.…”

Section: Methodsmentioning

confidence: 99%

Interpretable Sequence Classification via Discrete Optimization

Shvo

Icarte

et al. 2021

AAAI

View full text Add to dashboard Cite

Sequence classification is the task of predicting a class label given a sequence of observations. In many applications such as healthcare monitoring or intrusion detection, early classification is crucial to prompt intervention. In this work, we learn sequence classifiers that favour early classification from an evolving observation trace. While many state-of-the-art sequence classifiers are neural networks, and in particular LSTMs, our classifiers take the form of finite state automata and are learned via discrete optimization. Our automata-based classifiers are interpretable---supporting explanation, counterfactual reasoning, and human-in-the-loop modification---and have strong empirical performance. Experiments over a suite of goal recognition and behaviour classification datasets show our learned automata-based classifiers to have comparable test performance to LSTM-based classifiers, with the added advantage of being interpretable.

show abstract

Section: Methodsmentioning

confidence: 99%

Interpretable Sequence Classification via Discrete Optimization

Shvo

Icarte

et al. 2021

AAAI

View full text Add to dashboard Cite

show abstract

“…One example of interleaved plans in the computer game Minecraft would be an agent doing a plan to obtain bread while simultaneously also doing a plan to obtain potato (such as by gathering wheat and a potato before actually making the bread). Elexir-MCTS has demonstrated strong performance and improvement in scaling goal recognition over comprehensive search (Kantharaju et al, 2019); this recent success led us to use Elexir-MCTS in our study.…”

Section: Goal Recognition Via Combinatory Categorial Grammarsmentioning

confidence: 99%

“…Traces recording the behaviors of game-playing agents in Real-Time Strategy games have been used to recognize agents' goals (Kantharaju et al, 2019). Others have generated synthetic traces using a stochastic simulator in constrained domains (Ramírez and Geffner, 2009) and goal-directed agent in an open-world simulated environment (Rabkina et al, 2020).…”

Section: Inspectabilitymentioning

confidence: 99%

“…This method extends an existing approach for goal recognition via analogical reasoning, Analogical Theory of Mind (Rabkina et al, 2017(Rabkina et al, , 2020 by allowing it to leverage pre-existing goal recognition models. Second, we conduct an evaluation of RAGeR and two state-of-the-art goal recognition algorithms, PANDA-Rec (Höller et al, 2018) and Elexir-MCTS (Kantharaju et al, 2019), on data with varying inspectability and reliability from the open-world computer game Minecraft, and from the disaster management domain Monroe. We find that each approach to goal recognition has their own strengths and weaknesses.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of Goal Recognition Systems on Unreliable Data and Uninspectable Agents

Rabkina

Kantharaju

Wilson

et al. 2022

Front. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

Goal or intent recognition, where one agent recognizes the goals or intentions of another, can be a powerful tool for effective teamwork and improving interaction between agents. Such reasoning can be challenging to perform, however, because observations of an agent can be unreliable and, often, an agent does not have access to the reasoning processes and mental models of the other agent. Despite this difficulty, recent work has made great strides in addressing these challenges. In particular, two Artificial Intelligence (AI)-based approaches to goal recognition have recently been shown to perform well: goal recognition as planning, which reduces a goal recognition problem to the problem of plan generation; and Combinatory Categorical Grammars (CCGs), which treat goal recognition as a parsing problem. Additionally, new advances in cognitive science with respect to Theory of Mind reasoning have yielded an approach to goal recognition that leverages analogy in its decision making. However, there is still much unknown about the potential and limitations of these approaches, especially with respect to one another. Here, we present an extension of the analogical approach to a novel algorithm, Refinement via Analogy for Goal Reasoning (RAGeR). We compare RAGeR to two state-of-the-art approaches which use planning and CCGs for goal recognition, respectively, along two different axes: reliability of observations and inspectability of the other agent's mental model. Overall, we show that no approach dominates across all cases and discuss the relative strengths and weaknesses of these approaches. Scientists interested in goal recognition problems can use this knowledge as a guide to select the correct starting point for their specific domains and tasks.

show abstract

“…The choice to use plan-libraries for plan recognition stems from their high expressiveness, and their ability to provide robust explanations without any demonstrations or data sampling (Kim et al, 2018;Kantharaju et al, 2019). Within the scope of planlibrary based plan recognition, there is a variety of algorithms and representations, including AND/OR trees, grammars, Hierarchical Task Networks (HTNs) (Erol et al, 1995), Temporal Plan Networks (TPNs) Kim et al (2001), and more.…”

Section: Introductionmentioning

confidence: 99%

Comparing Plan Recognition Algorithms Through Standard Plan Libraries

Mirsky¹,

Galun²,

Gal³

et al. 2022

Front. Artif. Intell.

View full text Add to dashboard Cite

Plan recognition deals with reasoning about the goals and execution process of an actor, given observations of its actions. It is one of the fundamental problems of AI, applicable to many domains, from user interfaces to cyber-security. Despite the prevalence of these approaches, they lack a standard representation, and have not been compared using a common testbed. This paper provides a first step towards bridging this gap by providing a standard plan library representation that can be used by hierarchical, discrete-space plan recognition and evaluation criteria to consider when comparing plan recognition algorithms. This representation is comprehensive enough to describe a variety of known plan recognition problems and can be easily used by existing algorithms in this class. We use this common representation to thoroughly compare two known approaches, represented by two algorithms, SBR and Probabilistic Hostile Agent Task Tracker (PHATT). We provide meaningful insights about the differences and abilities of these algorithms, and evaluate these insights both theoretically and empirically. We show a tradeoff between expressiveness and efficiency: SBR is usually superior to PHATT in terms of computation time and space, but at the expense of functionality and representational compactness. We also show how different properties of the plan library affect the complexity of the recognition process, regardless of the concrete algorithm used. Lastly, we show how these insights can be used to form a new algorithm that outperforms existing approaches both in terms of expressiveness and efficiency.

show abstract

Scaling up CCG-Based Plan Recognition via Monte-Carlo Tree Search

Cited by 7 publications

References 9 publications

Interpretable Sequence Classification via Discrete Optimization

Interpretable Sequence Classification via Discrete Optimization

Evaluation of Goal Recognition Systems on Unreliable Data and Uninspectable Agents

Comparing Plan Recognition Algorithms Through Standard Plan Libraries

Contact Info

Product

Resources

About