Empowerment: A Universal Agent-Centric Measure of Control

Klyubin, A.S.; Polani, Daniel; Nehaniv, Chrystopher L.

doi:10.1109/cec.2005.1554676

Cited by 226 publications

(256 citation statements)

References 9 publications

Supporting

Mentioning

254

Contrasting

Order By: Relevance

“…The expected complexity or cost is exactly the same quantity minimized in risk-sensitive or KL control (Klyubin, Polani, & Nehaniv, 2005;van den Broek, Wiegerinck, & Kappen, 2010), and underpins related variational formulations of bounded rationality based on complexity costs (Braun, Ortega, Theodorou, & Schaal, 2011;Ortega & Braun, 2013). In other words, minimizing expected complexity renders behavior risk-sensitive, while maximizing expected accuracy induces ambiguity-sensitive behavior.…”

Section: Free Energy and Expected Free Energymentioning

confidence: 99%

Active Inference: A Process Theory

et al. 2017

View full text Add to dashboard Cite

This is the published version of the paper.This version of the publication may differ from the final published version. behavior prescribed by these dynamics has a degree of face validity, providing a formal explanation for reward seeking, context learning, and epistemic foraging. Technically, the fact that a gradient descent appears to be a valid description of neuronal activity means that variational free energy is a Lyapunov function for neuronal dynamics, which therefore conform to Hamilton's principle of least action. Permanent repository link

show abstract

Section: Free Energy and Expected Free Energymentioning

confidence: 99%

Active Inference: A Process Theory

et al. 2017

View full text Add to dashboard Cite

show abstract

“…These include empowerment Klyubin et al (2005), autonomy (Bertschinger et al, 2008), decisions (Tishby and Polani, 2011), and embodiment (Zahedi and Ay, 2013).…”

Section: Introductionmentioning

confidence: 99%

Action and perception for spatiotemporal patterns

Biehl¹,

Polani²

2017

Proceedings of the 14th European Conference on Artificial Life ECAL 2017

View full text Add to dashboard Cite

This is a contribution to the formalization of the concept of agents in multivariate Markov chains. Agents are commonly defined as entities that act, perceive, and are goal-directed. In a multivariate Markov chain (e.g. a cellular automaton) the transition matrix completely determines the dynamics. This seems to contradict the possibility of acting entities within such a system. Here we present definitions of actions and perceptions within multivariate Markov chains based on entitysets. Entity-sets represent a largely independent choice of a set of spatiotemporal patterns that are considered as all the entities within the Markov chain. For example, the entityset can be chosen according to operational closure conditions or complete specific integration. Importantly, the perceptionaction loop also induces an entity-set and is a multivariate Markov chain. We then show that our definition of actions leads to non-heteronomy and that of perceptions specialize to the usual concept of perception in the perception-action loop.

show abstract

“…They show that maximization of information flow can evolve into a meaningful sensorimotor structure (Klyubin et al 2004(Klyubin et al , 2007. In Klyubin et al (2005Klyubin et al ( , 2008) the authors present a universal agent-centric measure, called "empowerment", which is defined as the information-theoretic capacity of an agent's actuation channel (the maximum mutual information for the channel over all possible distributions of the transmitted signal). The empowerment is zero when the agent has no control over its sensory input, and it is higher when the agent can control what it is sensing.…”

Section: Information Flow In Adaptive Closed-loop Systemsmentioning

confidence: 99%

“…In the study of Klyubin et al (2004Klyubin et al ( , 2005Klyubin et al ( , 2007Klyubin et al ( , 2008) the authors used a Bayesian network to model perception-action loops. In their approach a perception-action loop is interpreted in terms of a communication channel-like model.…”

Section: Information Flow In Adaptive Closed-loop Systemsmentioning

confidence: 99%

Behavioral analysis of differential hebbian learning in closed-loop systems

et al. 2010

View full text Add to dashboard Cite

Understanding closed loop behavioral systems is a non-trivial problem, especially when they change during learning. Descriptions of closed loop systems in terms of information theory date back to the 1950s, however, there have been only a few attempts which take into account learning, mostly measuring information of inputs. In this study we analyze a specific type of closed loop system by looking at the input as well as the output space. For this, we investigate simulated agents that perform differential Hebbian learning (STDP). In the first part we show that analytical solutions can be found for the temporal development of such systems for relatively simple cases. In the second part of this study we try to answer the following question: How can we predict which system from a given class would be the best for a particular scenario? This question is addressed using energy, input/output ratio and entropy measures and investigating Tomas Kulvicius and Christoph Kolodziejski have contributed equally to this work. their development during learning. This way we can show that within well-specified scenarios there are indeed agents which are optimal with respect to their structure and adaptive properties. T. Kulvicius (B)·

show abstract

Empowerment: A Universal Agent-Centric Measure of Control

Cited by 226 publications

References 9 publications

Active Inference: A Process Theory

Active Inference: A Process Theory

Action and perception for spatiotemporal patterns

Behavioral analysis of differential hebbian learning in closed-loop systems

Contact Info

Product

Resources

About