Exploration in active learning

By Sebastian Thrun. 1998

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Information seeking, curiosity and attention: computational and neural mechanisms (2013, positive): “Therefore, heuristics based on novelty can guide efficient learning in small and closed spaces, where the number of tasks is small (Thrun 1995) but are very inefficient in large open ended spaces, where they only allow the agent to collect very sparse data and risk trapping him in unlearnable tasks ”
Active learning of inverse models with intrinsically motivated goal exploration in robots (2013, neutral): “en consisting in variants of expected informational gain. A large diversity of criteria can be used to evaluate the utility of given sampling candidates, such as the maximization of prediction errors [118], the local density of already queried points [131], the maximization of the decrease of global”
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016, neutral): “Thrun in [239] concluded that solving the exploration-exploitation dilemma, i.”
The challenges of joint attention (2006, background): “A major trend in these models is to study the behavior of systems driven by some sort of artificial curiosity or search for optimal experiments (Fedorov, 1972; Schmidhuber, 1991; Cohn et al., 1994; Thrun, 1995; Herrmann et al., 2000; Huang and Weng, 2002; Kaplan and Oudeyer, 2003; Kaplan and Oudey”
Intrinsically motivated goal exploration for active motor learning in robots: A case study (2010, background): “heuristics for maximizing prediction errors, variance, entropy or uncertainty [9]).”
Ieee Transactions on Evolutionary Computation Intrinsic Motivation Systems for Autonomous Mental Development (unknown): “This has shown to be efficient when the machine M has to learn a mapping which is learnable, deterministic and with homogeneous Gaussian noise ([32]; [15]; [17]; [21]).”
Guest Editorial Active Learning and Intrinsically Motivated Exploration in Robots: Advances and Challenges (2010, background): “A large diversity of criteria can be used to evaluate the utility of given sampling candidates, such as the maximization of prediction errors [22], the local density of already queried points [40], the maximization of the decrease of global model variance [2], expected improvement [11], or maximal…”
Structure-Adaptable Digital Neural Networks (1999, background): “Moreover, Thrun [209, 210] distinguishes between undirected and directed exploration.”
The interaction of maturational constraints and intrinsic motivations in active motor development (2011, background): “…of performance of the learning process to guide the exploration in the most " interesting " sensory-motor areas which maximize the notion of informational gain (e.g. heuristics for maximizing entropy, uncertainty, variance, prediction errors or decrease in prediction errors) [3], [4], [6].”
Active Learning and Intrinsically Motivated Exploration in Robots : Advances and Challenges (2010, background): “A large diversity of criteria can be used to evaluate the utility of given sampling candidates, such as the maximization of prediction errors [22], the local density of already queried points [40], the maximization of the decrease of global model variance [2], expected improvement [11], or maximal…”
The strategic student approach for life-long exploration and learning (2012, background): “uncertain [8], high prediction errors [17], least visited [18]) and then progressively explore what is more simple.”