Learning hierarchical task network domains from partially observed plan traces

Zhuo, Hankz Hankui; Muñoz-Ávila, Héctor; Yang, Qiang

doi:10.1016/j.artint.2014.04.003

Cited by 43 publications

(26 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We build planning problems and action models based on the analysis of the input music and dancing experts and calculate plans for robots to dance automatically using an off-the-shelf planner. In the future it would be interesting to study learning action models Zhuo, Muñoz-Avila, and Yang 2014) and planning with the learnt models (Zhuo and Kambhampati 2017) automatically from dances demonstrated by dancing experts.…”

Section: Resultsmentioning

confidence: 99%

Plan2Dance: Planning Based Choreographing from Music

Liu

Xie

Zhuo³

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

The field of dancing robots has drawn much attention from numerous sources. Despite the success of previous systems on choreography for robots to dance with external stimuli, they are often either limited to a pre-defined set of movements or lack of considering “hard” relations among dancing motions. In the demonstration, we design a planning based choreographing system, which views choreography with music as planning problems and solve the problems with off-the-shelf planners. Our demonstration exhibits the effectiveness of our system via evaluating our system with various music.

show abstract

Section: Resultsmentioning

confidence: 99%

Plan2Dance: Planning Based Choreographing from Music

Liu

Xie

Zhuo³

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

show abstract

“…The systems described in the previous paragraph behave that way. Another such system is HTNLearn (Zhuo, Muñoz-Avila, and Yang 2014), which cast the HTN learning problem as a constraint satisfaction problem such that when it is solved (e.g., by a constraint solver such as MAXSAT (Borchers and Furman 1998)), it will learn HTN methods. Our work is also non-incremental.…”

Section: Related Workmentioning

confidence: 99%

HTN Learning via Transfer Learning of Domain Landmarks

Pennisi

Fine-Morris

Floyd

et al. 2021

FLAIRS

View full text Add to dashboard Cite

Hierarchical Task Network (HTN) planning uses task-subtask relationships to break complex problems into more manageable subtasks, similar to how human problem-solvers plan. However, one limitation of HTN planning is that it requires domain knowledge in the form of planning methods to perform this task decomposition. Recent work has partially alleviated this knowledge engineering requirement by learning HTN methods from traces of observed behavior. Although this greatly reduces the amount of knowledge that must be encoded by a domain expert, it requires a large collection of traces in order to infer important landmark states that are used during trace segmentation and method learning. In this paper we present a novel method for landmark inference that transfers knowledge of landmarks from previously encountered environments to new environments without requiring any traces from the new environment. We evaluate our work in a logistics planning domain and show that our approach performs comparably to the existing landmark inference method but requires far fewer traces.

show abstract

“…A third strand of research that is also related to our work is that of action model learning. Work such as [53,61,60,57,56,55,52,13,18] focuses on learning action models directly from observed (or pre-specified) plan cases. The connection between this strand of work and our work can be seen in terms of the familiar up-front vs. demand-driven knowledge transfer: the learning methods attempt to condense the case library directly into STRIPS models before using it in planning, while we transfer knowledge from cases on a per-problem basis.…”

Section: Action Model Learningmentioning

confidence: 99%

Model-lite planning: Case-based vs. model-based approaches

Zhuo

Kambhampati

2017

Artificial Intelligence

Self Cite

View full text Add to dashboard Cite

There is increasing awareness in the planning community that depending on complete models impedes the applicability of planning technology in many real world domains where the burden of specifying complete domain models is too high. In this paper, we consider the problem of generating robust and accurate plans, when the agent only has access to incomplete domain models, supplanted by a set of successful plan cases. We will develop two classes of approaches-one case-based and the other modelbased. ML-CBP is a case-based approach that leverages the incomplete model and the plan cases to solve a new problem directly by affecting case-level transfer. RIM is a model-based approach that uses the incomplete model and the plan cases to first learn a more complete model. This model contains both primitive actions as well as macro-operators that are derived from the plan cases. The learned model is then used in conjunction with an off-the-shelf planner to solve new problems. We present a comprehensive evaluation of the two approaches, both to characterize their relative tradeoffs, and to quantify their advances over existing approaches.

show abstract

Learning hierarchical task network domains from partially observed plan traces

Cited by 43 publications

References 44 publications

Plan2Dance: Planning Based Choreographing from Music

Plan2Dance: Planning Based Choreographing from Music

HTN Learning via Transfer Learning of Domain Landmarks

Model-lite planning: Case-based vs. model-based approaches

Contact Info

Product

Resources

About