Learning and planning in partially observable environments without prior domain knowledge

2022 ◽  
Vol 142 ◽  
pp. 147-160
Author(s):  
Yunlong Liu ◽  
Jianyang Zheng ◽  
Fangfang Chang
Keyword(s):  
Domain Knowledge ◽  
Partially Observable

scholarly journals Adaptive Thompson Sampling Stacks for Memory Bounded Open-Loop Planning

2019 ◽  
Author(s):  
Thomy Phan ◽  
Thomas Gabor ◽  
Robert Müller ◽  
Christoph Roch ◽  
Claudia Linnhoff-Popien
Keyword(s):  
Domain Knowledge ◽  
Planning Horizon ◽  
Open Loop ◽  
Generative Model ◽  
Benchmark Problems ◽  
Adaptive Memory ◽  
Memory Consumption ◽  
Thompson Sampling ◽  
Planning Algorithms ◽  
Partially Observable

We propose Stable Yet Memory Bounded Open-Loop (SYMBOL) planning, a general memory bounded approach to partially observable open-loop planning. SYMBOL maintains an adaptive stack of Thompson Sampling bandits, whose size is bounded by the planning horizon and can be automatically adapted according to the underlying domain without any prior domain knowledge beyond a generative model. We empirically test SYMBOL in four large POMDP benchmark problems to demonstrate its effectiveness and robustness w.r.t. the choice of hyperparameters and evaluate its adaptive memory consumption. We also compare its performance with other open-loop planning algorithms and POMCP.

scholarly journals Multimodal estimation and communication of latent semantic knowledge for robust execution of robot instructions

2020 ◽  
Vol 39 (10-11) ◽  
pp. 1279-1304 ◽  
Author(s):  
Jacob Arkin ◽  
Daehyung Park ◽  
Subhro Roy ◽  
Matthew R Walter ◽  
Nicholas Roy ◽  
...  
Keyword(s):  
Domain Knowledge ◽  
Knowledge Bases ◽  
Semantic Knowledge ◽  
Linguistic Knowledge ◽  
World Knowledge ◽  
World State ◽  
Contradictory Evidence ◽  
Partially Observable ◽  
Visual Properties ◽  
Semantic Properties

The goal of this article is to enable robots to perform robust task execution following human instructions in partially observable environments. A robot’s ability to interpret and execute commands is fundamentally tied to its semantic world knowledge. Commonly, robots use exteroceptive sensors, such as cameras or LiDAR, to detect entities in the workspace and infer their visual properties and spatial relationships. However, semantic world properties are often visually imperceptible. We posit the use of non-exteroceptive modalities including physical proprioception, factual descriptions, and domain knowledge as mechanisms for inferring semantic properties of objects. We introduce a probabilistic model that fuses linguistic knowledge with visual and haptic observations into a cumulative belief over latent world attributes to infer the meaning of instructions and execute the instructed tasks in a manner robust to erroneous, noisy, or contradictory evidence. In addition, we provide a method that allows the robot to communicate knowledge dissonance back to the human as a means of correcting errors in the operator’s world model. Finally, we propose an efficient framework that anticipates possible linguistic interactions and infers the associated groundings for the current world state, thereby bootstrapping both language understanding and generation. We present experiments on manipulators for tasks that require inference over partially observed semantic properties, and evaluate our framework’s ability to exploit expressed information and knowledge bases to facilitate convergence, and generate statements to correct declared facts that were observed to be inconsistent with the robot’s estimate of object properties.

scholarly journals A Decentralized Partially Observable Decision Model for Recognizing the Multiagent Goal in Simulation Systems

2016 ◽  
Vol 2016 ◽  
pp. 1-15 ◽  
Author(s):  
Shiguang Yue ◽  
Kristina Yordanova ◽  
Frank Krüger ◽  
Thomas Kirste ◽  
Yabing Zha
Keyword(s):  
Domain Knowledge ◽  
Decision Model ◽  
Training Dataset ◽  
Predator Prey ◽  
Modeling Methods ◽  
Model Free ◽  
Goal Recognition ◽  
Cooperative Behaviors ◽  
Partially Observable ◽  
Simulation Systems

Multiagent goal recognition is important in many simulation systems. Many of the existing modeling methods need detailed domain knowledge of agents’ cooperative behaviors and a training dataset to estimate policies. To solve these problems, we propose a novel decentralized partially observable decision model (Dec-POMDM), which models cooperative behaviors by joint policies. In this compact way, we only focus on the distribution of joint policies. Additionally, a model-free algorithm, cooperative colearning based on Sarsa, is exploited to estimate agents’ policies under the assumption of rationality, which makes the training dataset unnecessary. In the inference, considering that the Dec-POMDM is discrete and its state space is large, we implement a marginal filter (MF) under the framework of the Dec-POMDM, where the initial world states and results of actions are uncertain. In the experiments, a new scenario is designed based on the standard predator-prey problem: we increase the number of preys, and our aim is to recognize the real target of predators. Experiment results show that (a) our method recognizes goals well even when they change dynamically; (b) the Dec-POMDM outperforms supervised trained HMMs in terms of precision, recall, and F-measure; and (c) the MF infers goals more efficiently than the particle filter under the framework of the Dec-POMDM.

Automated Assessment of Domain Knowledge With Online Knowledge Mapping

2006 ◽  
Author(s):  
Gregory K. W. K. Chung ◽  
Eva L. Baker ◽  
David G. Brill ◽  
Ravi Sinha ◽  
Farzad Saadat ◽  
...  
Keyword(s):  
Domain Knowledge ◽  
Knowledge Mapping ◽  
Automated Assessment

A Model for Structured Data Entry Based on Explicit Descriptional Knowledge

1994 ◽  
Vol 33 (05) ◽  
pp. 454-463 ◽  
Author(s):  
A. M. van Ginneken ◽  
J. van der Lei ◽  
J. H. van Bemmel ◽  
P. W. Moorman
Keyword(s):  
Domain Knowledge ◽  
Data Entry ◽  
Knowledge Bases ◽  
Structured Data ◽  
Research Quality ◽  
Free Text ◽  
Specific Knowledge ◽  
User Input ◽  
The One ◽  
Research Quality Assessment

Abstract:Clinical narratives in patient records are usually recorded in free text, limiting the use of this information for research, quality assessment, and decision support. This study focuses on the capture of clinical narratives in a structured format by supporting physicians with structured data entry (SDE). We analyzed and made explicit which requirements SDE should meet to be acceptable for the physician on the one hand, and generate unambiguous patient data on the other. Starting from these requirements, we found that in order to support SDE, the knowledge on which it is based needs to be made explicit: we refer to this knowledge as descriptional knowledge. We articulate the nature of this knowledge, and propose a model in which it can be formally represented. The model allows the construction of specific knowledge bases, each representing the knowledge needed to support SDE within a circumscribed domain. Data entry is made possible through a general entry program, of which the behavior is determined by a combination of user input and the content of the applicable domain knowledge base. We clarify how descriptional knowledge is represented, modeled, and used for data entry to achieve SDE, which meets the proposed requirements.

Classification of means and methods of the Web semantic retrieval

2017 ◽  
pp. 030-050
Author(s):  
J.V. Rogushina ◽  
Keyword(s):  
Search Engines ◽  
Domain Knowledge ◽  
Information Needs ◽  
Web Search ◽  
User Interaction ◽  
Query Languages ◽  
Semantic Search ◽  
Semantic Retrieval ◽  
The Web

Problems associated with the improve ment of information retrieval for open environment are considered and the need for it’s semantization is grounded. Thecurrent state and prospects of development of semantic search engines that are focused on the Web information resources processing are analysed, the criteria for the classification of such systems are reviewed. In this analysis the significant attention is paid to the semantic search use of ontologies that contain knowledge about the subject area and the search users. The sources of ontological knowledge and methods of their processing for the improvement of the search procedures are considered. Examples of semantic search systems that use structured query languages (eg, SPARQL), lists of keywords and queries in natural language are proposed. Such criteria for the classification of semantic search engines like architecture, coupling, transparency, user context, modification requests, ontology structure, etc. are considered. Different ways of support of semantic and otology based modification of user queries that improve the completeness and accuracy of the search are analyzed. On base of analysis of the properties of existing semantic search engines in terms of these criteria, the areas for further improvement of these systems are selected: the development of metasearch systems, semantic modification of user requests, the determination of an user-acceptable transparency level of the search procedures, flexibility of domain knowledge management tools, increasing productivity and scalability. In addition, the development of means of semantic Web search needs in use of some external knowledge base which contains knowledge about the domain of user information needs, and in providing the users with the ability to independent selection of knowledge that is used in the search process. There is necessary to take into account the history of user interaction with the retrieval system and the search context for personalization of the query results and their ordering in accordance with the user information needs. All these aspects were taken into account in the design and implementation of semantic search engine "MAIPS" that is based on an ontological model of users and resources cooperation into the Web.

Modeling Affordances and Functioning for Personalized Robotic Assistance

2020 ◽  
Author(s):  
Alessandro Umbrico ◽  
Gabriella Cortellessa ◽  
Andrea Orlandini ◽  
Amedeo Cesta
Keyword(s):  
Domain Knowledge ◽  
Experimental Tests ◽  
International Classification Of Functioning ◽  
Knowledge Representation And Reasoning ◽  
Health State ◽  
Knowledge Reasoning ◽  
Cognitive Assistance ◽  
Ontological Model ◽  
And Function

A key aspect of robotic assistants is their ability to contextualize their behavior according to different needs of assistive scenarios. This work presents an ontology-based knowledge representation and reasoning approach supporting the synthesis of personalized behavior of robotic assistants. It introduces an ontological model of health state and functioning of persons based on the International Classification of Functioning, Disability and Health. Moreover, it borrows the concepts of affordance and function from the literature of robotics and manufacturing and adapts them to robotic (physical and cognitive) assistance domain. Knowledge reasoning mechanisms are developed on top of the resulting ontological model to reason about stimulation capabilities of a robot and health state of a person in order to identify action opportunities and achieve personalized assistance. Experimental tests assess the performance of the proposed approach and its capability of dealing with different profiles and stimuli.

Identifying The Competencies Possessed By The Engineering College Faculty In India

2019 ◽  
Vol 118 (4) ◽  
pp. 68-71
Author(s):  
Mohanakumari. D ◽  
R. Magesh
Keyword(s):  
Domain Knowledge ◽  
Teaching And Learning ◽  
Vital Role ◽  
Faculty Members ◽  
Approaches To Teaching ◽  
Global Workforce ◽  
Content Domain ◽  
Technology Changes ◽  
Education In India ◽  
Pre Service Teacher

The main intention of the Paper is identifying the competencies possessed by the faculty in engineering college and adequate skills of all the disciplines required and that plays a vital role in educational institutions.In this era, engineering education in India faces major challenges as it requires meeting the demands of technical profession and emerging job market. Researchers have created some universally desired, yet challenging skills for global workforce. Nowadays, technology changes rapidly, so we have to update our self-according to the changing world, i.e., infrastructure, content/domain knowledge, educators/HR trainers. Thus, our technical faculty members should necessary to learn the innovative approaches to teaching and learning, which in turn will require effective professional development for both new and experienced instructors alike. It is right time now to redesign our curriculam, pedagogy and make the pre-service teacher preparation programme mandatory part of technical higher education.

Sign in / Sign up

Export Citation Format

Share Document