Active Learning and Intrinsically Motivated Exploration in Robots: Advances and Challenges

Active Learning and Intrinsically Motivated Exploration in Robots: Advances and Challenges (2010)

Cached

Download Links

by Manuel Lopes, et al.

Citations:

BibTeX

@MISC{Lopes10activelearning,
    author = {Manuel Lopes and et al.},
    title = { Active Learning and Intrinsically Motivated Exploration in Robots: Advances and Challenges},
    year = {2010}
}

Bookmark

OpenURL

Abstract

LEARNING techniques are increasingly being used in today’s complex robotic systems. Robots are expected to deal with a large variety of tasks using their high-dimensional and complex bodies, to manipulate objects and also, to interact with humans in an intuitive and friendly way. In this new setting, not all relevant information is available at design time, and robots should typically be able to learn, through self-experimentation or through human–robot interaction, how to tune their innate perceptual-motor skills or to learn, cumulatively, novel skills that were not preprogrammed initially. In a word, robots need to have the capacity to develop in an open-ended manner and in an open-ended environment, in a way that is analogous to human development which combines genetic and epigenetic factors. This challenge is at the center of the developmental robotics field [7], [35]–[37]. Among the various

Citations

582	Queries and concept learning - Angluin - 1988
427	Integrated architectures for learning, planning, and reacting based on approximating integrated architectures for learning, planning, and reacting based on approximating dynamic programming - Sutton - 1990
402	Active learning with statistical models - Cohn, Ghahramani, et al. - 1996
322	PR: A neural substrate of prediction and reward - Schultz, Dayan, et al. - 1997
258	Theory of Optimal Experiments - Fedorov - 1972
161	is more: active learning with support vector machines - Schohn, Cohn, et al. - 2000
123	Autonomous mental development by robots and animals - Weng, McClelland, et al. - 2001
121	Efficient global optimization of expensive black-box functions, Global Optimization 13 - Jones, Schonlau, et al. - 1998
111	Bayesian experimental design: A review - Chaloner, Verdinelli - 1995
91	Near-optimal sensor placements in Gaussian processes: theory, efficientalgorithms and empirical studies - Krause, Singh, et al. - 2008
81	Intrinsic motivation systems for autonomous mental development - Oudeyer, Kaplan, et al. - 2007
80	Intrinsically motivated learning of hierarchical collections of skills - Barto, Singh, et al. - 2004
76	Developmental robotics: A survey - Lungarella, Metta, et al. - 2003
68	Motivation reconsidered: The concept of competence - White - 1959
68	Monotone Dynamical Systems - Smith - 1995
62	Confl ict, Arousal and Curiosity - Berlyne - 1960
59	Intrinsically motivated reinforcement learning - Singh, Barto, et al. - 2005
58	Active Exploration in Dynamic Environments - Thrun, Moeller - 1992
57	A possibility for implementing curiosity and boredom in model-building neural controllers - Schmidhuber - 1991
55	How to build complete creatures rather than isolated cognitive simulators - Brooks - 1991
50	Analysis of a greedy active learning strategy - Dasgupta - 2004
46	Novelty and reinforcement learning in the value system of developmental robots - Huang, Weng - 2002
35	Interactive policy learning through confidence-based autonomy - Chernova, Veloso
31	Minimax bounds for active learning - Castro, Nowak - 2007
30	The short-latency dopamine signal: a role in discovering novel actions? Nature Reviews Neuroscience 7 - Redgrave, Gurney - 2006
28	A study of cooperative mechanisms for faster reinforcement learning - Whitehead - 1991
27	Intrinsic Motivation and Self-Determination - Deci, Ryan - 1985
26	The playground experiment: taskindependent development of a curious robot - Oudeyer, Kaplan, et al.
26	Exploration in active learning - Thrun - 1995
26	Dopamine: Generalization and bonuses - Kakade, Dayan - 2002
24	The true sample complexity of active learning - Balcan, Hanneke, et al.
24	Bringing up robot: Fundamental mechanisms for creating a self-motivated, self-organizing architecture - Blank, Kumar, et al. - 2002
22	Active learning with Gaussian processes for object categorization - Kapoor, Grauman, et al. - 2007
20	Compressed sensing and Bayesian experimental design - Seeger, Nickisch
18	Developmental robotics, optimal artificial curiosity, creativity, music, and the fine arts - Schmidhuber - 2006
17	Intrinsically motivated hierarchical manipulation - Hart, Grupen - 2008
16	Nonmyopic active learning of Gaussian processes: an exploration–exploitation approach - Krause, Guestrin - 2008
15	How can we define intrinsic motivations - Oudeyer, Kaplan - 2008
15	Active learning for logistic regression: an evaluation - Schein, Ungar - 2007
12	Cognitive developmental robotics: A survey - Asada, Hosoda, et al. - 2009
11	Riac: Robust intrinsically motivated exploration and active learning - Baranes, Oudeyer - 2009
10	A Bayesian exploration-exploitation approach for optimal online sensing and planning with a visually guided mobile robot. Autonomous Robots - Martinez-Cantin, Freitas, et al. - 2009
10	Intrinsically motivated reinforcement learning: An evolutionary perspective - Singh, Lewis, et al. - 2010
10	2007b). Evolution and learning in an intrinsically motivated reinforcement learning robot - Schembri, Mirolli, et al.
9	Active learning for reward estimation in inverse reinforcement learning - Lopes, Melo, et al. - 2009
8	Generalised weakened fictitious play - Leslie, Collins - 2006
5	Thomaz, “Designing interactions for robot active learners - Cakmak, Chao, et al. - 2010
5	Learning invariant sensorimotor behaviors: A developmental approach to imitation mechanisms - Andry, Gaussier, et al. - 2003
4	Body schema acquisition through active learning - Cantin-Martinez, Lopes, et al. - 2010
4	Genetic programming for reward function search - Niekum, Barto, et al. - 2010