Filters








2,937 Hits in 4.1 sec

Introduction to the Special Issue on Machine Learning for Multiple Modalities in Interactive Systems and Robots

Heriberto Cuayáhuitl, Lutz Frommberger, Nina Dethlefs, Antoine Raux, Mathew Marge, Hendrik Zender
2014 ACM transactions on interactive intelligent systems (TiiS)  
For example, a robot may coordinate its speech with its actions, taking into account (audio-) visual feedback during their execution.  ...  However, machine learning methods that encompass multiple modalities of an interactive system are still relatively hard to find.  ...  ACKNOWLEDGMENTS We thank the chief editors of ACM TiiS, Anthony Jameson, John Riedl and Krzysztof Gajos, for letting this special issue become a reality and for their high commitment with this journal.  ... 
doi:10.1145/2670539 fatcat:mpwlonu2yfcnher33owzkl6j6y

How smart are our environments? An updated look at the state of the art

Diane J. Cook, Sajal K. Das
2007 Pervasive and Mobile Computing  
We also discuss ongoing challenges for continued research.  ...  Assuming a cooperative environment, they proposed [75] a cooperative game theory based learning policy for location-aware resource management in multi-inhabitant smart homes.  ...  However, unlike the reinforcement learner approaches, automation is based on imitating inhabitant behavior and therefore is more difficult to employ for alternative goals such as energy efficiency.  ... 
doi:10.1016/j.pmcj.2006.12.001 fatcat:2iwbmer3rfh2liwpzfxzpco5ae

ConvLab: Multi-Domain End-to-End Dialog System Platform

Sungjin Lee, Qi Zhu, Ryuichi Takanobu, Zheng Zhang, Yaoqin Zhang, Xiang Li, Jinchao Li, Baolin Peng, Xiujun Li, Minlie Huang, Jianfeng Gao
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: System Demonstrations  
showcase, we extend the MultiWOZ dataset with user dialog act annotations to train all component models and demonstrate how ConvLab makes it easy and effortless to conduct complicated experiments in multi-domain  ...  Multi-task learning An agent can have multiple bodies in different environments for the sake of transfer learning.  ...  as well as hierarchical reinforcement learning approaches (Peng et al., 2017) .  ... 
doi:10.18653/v1/p19-3011 dblp:conf/acl/LeeZTZZLLPLHG19 fatcat:3ddyloxlrfeejh6en5dq7od4xm

ConvLab: Multi-Domain End-to-End Dialog System Platform [article]

Sungjin Lee, Qi Zhu, Ryuichi Takanobu, Xiang Li, Yaoqin Zhang, Zheng Zhang, Jinchao Li, Baolin Peng, Xiujun Li, Minlie Huang, Jianfeng Gao
2019 arXiv   pre-print
We present ConvLab, an open-source multi-domain end-to-end dialog system platform, that enables researchers to quickly set up experiments with reusable components and compare a large set of different approaches  ...  showcase, we extend the MultiWOZ dataset with user dialog act annotations to train all component models and demonstrate how ConvLab makes it easy and effortless to conduct complicated experiments in multi-domain  ...  The multi-domain end-to-end task completion dialog track in DSTC8 will employ ConvLab as the challenge platform, giving rise to a reference use case.  ... 
arXiv:1904.08637v1 fatcat:6slmzkdo7vctzkojxdqapc7er4

Efficient Computation Offloading in Edge Computing Enabled Smart Home

Bocheng Yu, Xingjun Zhang, Ilsun You, Umer Sadiq Khan
2021 IEEE Access  
Next, we train a deep learning model for imitation branching policy.  ...  To the best of our knowledge, it is the first time to study the problem of cost and energy efficiency in a multi-layer network for smart home.  ... 
doi:10.1109/access.2021.3066789 fatcat:irhla7zrbbantbzbgwwbkrfihu

From Motor Learning to Interaction Learning in Robots [chapter]

Olivier Sigaud, Jan Peters
2010 Studies in Computational Intelligence  
For fast learning of the motor tasks, imitation learning offers the most promising approach. Self improvement requires reinforcement learning approaches that scale into the domain of complex robots.  ...  We focus here on several core domains of robot learning. For accurate task execution, we need motor learning capabilities.  ...  Motor learning, Imitation learning and Interaction Learning Making humanoid robots perform movements of the same agility as human movements is an aim difficult to achieve even for the simplest tasks.  ... 
doi:10.1007/978-3-642-05181-4_1 fatcat:zpfcyzaa7ra4xplm2hrxpnjtru

Behavior-based robotics as a tool for synthesis of artificial behavior and analysis of natural behavior

Maja J Matarić
1998 Trends in Cognitive Sciences  
For example, the 'avoid-obstacles' behavior maintains the goal of preventing collisions with objects in the environment, and the 'go-home' behavior achieves the goal of reaching some home region.  ...  Matarić is at the Review M a t a r ić -B e h a v i o r -b a s e d r o b o t i c s Reinforcement learning in the multi-robot domain Autonomous Robots 4, 73-83 6 Steels, L. (1994) The artificial life roots  ...  The author is grateful to Paolo Gaudiano for his extensive insightful comments on an earlier draft of this paper.  ... 
doi:10.1016/s1364-6613(98)01141-3 pmid:21227083 fatcat:sd3gf4aj2zbvrlzte4i7mcwg5e

Potential Impacts of Smart Homes on Human Behavior: A Reinforcement Learning Approach [article]

Shashi Suman, Ali Etemad, Francois Rivest
2021 arXiv   pre-print
The hierarchical human model learns to complete each activity and set optimal thermal settings for maximum comfort.  ...  We design a semi-Markov decision process human task interleaving model based on hierarchical reinforcement learning that learns to make decisions to either pursue or leave an activity.  ...  Our smart home is based on Q-learning and learns the preferences of the human model for each activity.  ... 
arXiv:2102.13307v3 fatcat:ixgtbkmhpfanhof64byjxf7zcm

Deep Reinforcement Learning [article]

Yuxi Li
2018 arXiv   pre-print
Then we discuss important mechanisms for RL, including attention and memory, unsupervised learning, hierarchical RL, multi-agent RL, relational RL, and learning to learn.  ...  We start with background of artificial intelligence, machine learning, deep learning, and reinforcement learning (RL), with resources.  ...  Lanctot et al. (2017) observe that independent RL, in which each agent learns by interacting with the environment, oblivious to other agents, can overfit the learned policies to other agents' policies  ... 
arXiv:1810.06339v1 fatcat:kp7atz5pdbeqta352e6b3nmuhy

Deep Reinforcement Learning: An Overview [article]

Yuxi Li
2018 arXiv   pre-print
After that, we discuss important mechanisms for RL, including attention and memory, unsupervised learning, transfer learning, multi-agent RL, hierarchical RL, and learning to learn.  ...  Please see Deep Reinforcement Learning, arXiv:1810.06339, for a significant update.  ...  See Ruder (2017) for an overview about multi-task learning.  ... 
arXiv:1701.07274v6 fatcat:x2es3yf3crhqblbbskhxelxf2q

Operation Mode Decision of Indoor Cleaning Robot Based on Causal Reasoning and Attribute Learning (February 2020)

Yapeng Li, Dongbo Zhang, Feng Yin, Ying Zhang
2020 IEEE Access  
INDEX TERMS Attributes learning, causal reasoning, cleaning robot, fuzzy inference network, joint learning.  ...  The method proposed in this paper imitates the way that human dispose of different types of garbage and has good interpretability.  ...  TABLE 3 . 3 Confusion matrix for Single task mode attribute learning. TABLE 4 . 4 Confusion matrix for ungrouped multi-task attribute.  ... 
doi:10.1109/access.2020.3003343 fatcat:7wdq5ykt4zeodpwmqlcdtvqwje

From Humans and Back: a Survey on Using Machine Learning to both Socially Perceive Humans and Explain to Them Robot Behaviours

Adina M. Panchea, François Ferland
2020 Current Robotics Reports  
There are papers which report models for robots to imitate humans and also for humans to imitate robots.  ...  To do so, machine learning (ML) is often employed.  ...  [59] , the use of bidirectional communication between a robot and a human is introduced to interactively learn hierarchical tasks from demonstrations.  ... 
doi:10.1007/s43154-020-00013-6 fatcat:l5dneve33faolgvlvf3ddqkv24

Learning Cloth Folding Tasks with Refined Flow Based Spatio-Temporal Graphs [article]

Peng Zhou, Omar Zahra, Anqing Duan, Shengzeng Huo, Zeyu Wu, David Navarro-Alarcon
2021 arXiv   pre-print
Cloth folding is a widespread domestic task that is seemingly performed by humans but which is highly challenging for autonomous robots to execute due to the highly deformable nature of textiles; It is  ...  In this paper, we propose a new solution for robotic cloth folding (using a standard folding board) via learning from demonstrations.  ...  To solve imitation learning for cloth folding tasks, we propose a hierarchical encoding for dominant motion representations extracted from a single demonstration video, called a refined optical flow-based  ... 
arXiv:2110.08620v1 fatcat:423mipzvwzghdnnhby66vjbfrm

Imitating Interactive Intelligence [article]

Josh Abramson, Arun Ahuja, Iain Barr, Arthur Brussee, Federico Carnevale, Mary Cassin, Rachita Chhaparia, Stephen Clark, Bogdan Damoc, Andrew Dudzik, Petko Georgiev, Aurelia Guy (+17 others)
2021 arXiv   pre-print
Therefore, we approximate the role of the human with another learned agent, and use ideas from inverse reinforcement learning to reduce the disparities between human-human and agent-agent interactive behaviour  ...  Taken together, our results in this virtual environment provide evidence that large-scale human behavioural imitation is a promising tool to create intelligent, interactive agents, and the challenge of  ...  Rezende, and others for important discussions.  ... 
arXiv:2012.05672v2 fatcat:2jeh6widnve55ozlxjunmdi24y

A Survey of Autonomous Human Affect Detection Methods for Social Robots Engaged in Natural HRI

Derek McColl, Alexander Hong, Naoaki Hatakeyama, Goldie Nejat, Beno Benhabib
2015 Journal of Intelligent and Robotic Systems  
This survey paper presents an encompassing review of existing automated affect recognition and classification systems for social robots engaged in various HRI settings.  ...  The automated systems are described by their corresponding robotic and HRI applications, the sensors they employ, and the feature detection techniques and affect classification strategies utilized.  ...  They have been proposed for or applied to different HRI tasks, including collaboration, assistance, mimicry, as well as multi-purpose HRI scenarios.  ... 
doi:10.1007/s10846-015-0259-2 fatcat:szdkrmfx7jf3fhfeilfpirxhku
« Previous Showing results 1 — 15 out of 2,937 results