1) La descarga del recurso depende de la página de origen
2) Para poder descargar el recurso, es necesario ser usuario registrado en Universia


Opción 1: Descargar recurso

Detalles del recurso

Descripción

Social interacting is a complex task for which machine learning holds particular promise. However, as no sufficiently accurate simulator of human interactions exists today, the learning of social interaction strategies has to happen online in the real world. Actions executed by the robot impact on humans, and as such have to be carefully selected, making it impossible to rely on random exploration. Additionally, no clear reward function exists for social interactions. This implies that traditional approaches used for Reinforcement Learning cannot be directly applied for learning how to interact with the social world. As such we argue that robots will profit from human expertise and guidance to learn social interactions. However, as the quantity of input a human can provide is limited, new methods have to be designed to use human input more efficiently. In this paper we describe a setup in which we combine a framework called Supervised Progressively Autonomous Robot Competencies (SPARC), which allows safer online learning with Reinforcement Learning, with the use of partial states rather than full states to accelerate generalisation and obtain a usable action policy more quickly.

Pertenece a

Faculty of Technology ePrints Service  

Autor(es)

Senft, Emmanuel -  Lemaignan, Severin -  Baxter, Paul -  Belpaeme, Tony - 

Id.: 70961932

Idioma: inglés  - 

Versión: 1.0

Estado: Final

Tipo:  application/pdf - 

Palabras claveG400 Computer Science - 

Tipo de recurso: Conference or Workshop contribution  -  NonPeerReviewed  - 

Tipo de Interactividad: Expositivo

Nivel de Interactividad: muy bajo

Audiencia: Estudiante  -  Profesor  -  Autor  - 

Estructura: Atomic

Coste: no

Copyright: sí

Formatos:  application/pdf - 

Requerimientos técnicos:  Browser: Any - 

Relación: [References] http://eprints.lincoln.ac.uk/30193/
[References] https://aaai.org/ocs/index.php/FSS/FSS17/paper/view/16011/15296

Fecha de contribución: 04-feb-2018

Contacto:

Localización:
* Senft, Emmanuel and Lemaignan, Severin and Baxter, Paul and Belpaeme, Tony (2017) Toward supervised reinforcement learning with partial states for social HRI. In: 4th AAAI FSS on Artificial Intelligence for Social Human-Robot Interaction (AI-HRI), 9 - 11 November 2017, Arlington, Virginia.

Otros recursos del mismo autor(es)

  1. Leveraging human inputs in interactive machine learning for human robot interaction A key challenge of HRI is allowing robots to be adaptable, especially as robots are expected to pene...
  2. From characterising three years of HRI to methodology and reporting recommendations Human-Robot Interaction (HRI) research requires the integration and cooperation of multiple discipli...
  3. Solve memory to solve cognition The foundations of cognition and cognitive behaviour are consistently proposed to be built upon the ...
  4. Simulation and HRI Recent Perspectives with the MORSE Simulator International audience
  5. Networking Needs and Solutions for Road Vehicles at Imara International audience

Otros recursos de la mismacolección

  1. An inhuman art Realisms and Object Orientations: Art, Politics and the philosophy of Tristan Garcia This multidisc...
  2. Science-art: accelerated models of spectating My paper proposes a role for contemporary art as part of the broader accelerationist project, summar...
  3. The nemocentric spectator I shall present for thirty minutes on a recent research project, in which I take inspiration from th...
  4. Realist spectating: time present I am curious about the way that some recent art deploys scientific discourse as a component part. I ...
  5. Science-art: neuronaturalism beyond the decentred spectator Science-art collaborations are a growth area. An example is the Wellcome Collection exhibition “Stat...

Aviso de cookies: Usamos cookies propias y de terceros para mejorar nuestros servicios, para análisis estadístico y para mostrarle publicidad. Si continua navegando consideramos que acepta su uso en los términos establecidos en la Política de cookies.