Learning to generate articulated behavior through the bottom-up and the top-down interaction processes

Tani, Jun

doi:10.1016/s0893-6080(02)00214-9

Cited by 136 publications

(111 citation statements)

References 13 publications

Supporting

Mentioning

110

Contrasting

Order By: Relevance

“…This is because there are prior beliefs that can reproduce an optimal policy to minimise expected cost. However, there are many other prior beliefs that specify Bayes-optimal control that do not minimise expected cost-see the handwriting simulations in Friston et al (2011) or the animate behaviours in Tani (2003).…”

Section: Perception and Actionmentioning

confidence: 99%

See 1 more Smart Citation

Active inference and agency: optimal control without cost functions

2012

View full text Add to dashboard Cite

This paper describes a variational free-energy formulation of (partially observable) Markov decision problems in decision making under uncertainty. We show that optimal control can be cast as active inference. In active inference, both action and posterior beliefs about hidden states minimise a free energy bound on the negative log-likelihood of observed states, under a generative model. In this setting, reward or cost functions are absorbed into prior beliefs about state transitions and terminal states. Effectively, this converts optimal control into a pure inference problem, enabling the application of standard Bayesian filtering techniques. We then consider optimal trajectories that rest on posterior beliefs about hidden states in the future. Crucially, this entails modelling control as a hidden state that endows the generative model with a representation of agency. This leads to a distinction between models with and without inference on hidden control states; namely, agency-free and agency-based models, respectively.

show abstract

Section: Perception and Actionmentioning

confidence: 99%

“…In robotics and engineering, the equivalent learning requires the agent to be shown how to perform a task. This form of learning has been used to produce some compelling and animate behaviours (Tani 2003;Namikawa et al 2011).…”

Section: Introductionmentioning

confidence: 99%

Active inference and agency: optimal control without cost functions

2012

View full text Add to dashboard Cite

show abstract

“…Modulation of the control neurons' activities causes shifts between generating one primitive and another. The scheme is analogous to the idea of the parametric bias [7] and the command neuron concept ( [8], [9], [10]). How might more complex tasks, such as navigation in an environment, be generated?…”

Section: Methodsmentioning

confidence: 99%

Evolved Motor Primitives and Sequences in a Hierarchical Recurrent Neural Network

Paine

Tani

2004

Genetic and Evolutionary Computation – GECCO 2004

Self Cite

View full text Add to dashboard Cite

Abstract. This study describes how complex goal-directed behavior can evolve in a hierarchically organized recurrent neural network controlling a simulated Khepera robot. Different types of dynamic structures self-organize in the lower and higher levels of a network for the purpose of achieving complex navigation tasks. The parametric bifurcation structures that appear in the lower level explain the mechanism of how behavior primitives are switched in a top-down way. In the higher level, a topologically ordered mapping of initial cell activation states to motor-primitive sequences self-organizes by utilizing the initial sensitivity characteristics of nonlinear dynamical systems. A further experiment tests the evolved controller's adaptability to changes in its environment. The biological plausibility of the model's essential principles is discussed.

show abstract

“…In other words, different dynamics need to be described by different connectionist models. To deal with a complex system containing multiple attractor dynamics, Tani et al proposed a RNN with parametric bias (RNNPB) (5)- (8) , and the novel RNN has been widely applied to behavior learning of robots recently (9) (10) (11) .…”

Section: Introductionmentioning

confidence: 99%

A Modified Recurrent Neural Network with Parametric Bias and its Application to Action Learning of a Humanoid Robot

Kuremoto

Morisaki

Kobayashi³

et al. 2014

Proceedings of the 2nd International Conference on Intelligent Systems and Image Processing 2014

View full text Add to dashboard Cite

Tani et al.'s recurrent neural network with parametric bias (RNNPB) is able to learning different time series patterns, and it has been successfully applied in action learning of robots. In this paper, we propose a novel type RNNPB using Elman type model instead of Jordan type in the conventional model. The proposed structure makes it easy to use error back-propagation (BP) learning algorithm which has lower computational cost than back-propagation time through (BPTT) method used in Tani et al.'s model. The effectiveness of the modified RNNPB was confirmed by its application to gesture learning experiment using a humanoid robot.

show abstract

Learning to generate articulated behavior through the bottom-up and the top-down interaction processes

Cited by 136 publications

References 13 publications

Active inference and agency: optimal control without cost functions

Active inference and agency: optimal control without cost functions

Evolved Motor Primitives and Sequences in a Hierarchical Recurrent Neural Network

A Modified Recurrent Neural Network with Parametric Bias and its Application to Action Learning of a Humanoid Robot

Contact Info

Product

Resources

About