Book ChapterDOI
Self-exploration of the Stumpy Robot with Predictive Information Maximization
Georg Martius,Luisa Jahn,Luisa Jahn,Helmut Hauser,Verena V. Hafner +4 more
- pp 32-42
TLDR
This work studies an intrinsic motivation system for behavioral self-exploration based on the maximization of the predictive information using the Stumpy robot, which is the first evaluation of the algorithm on a real robot.Abstract:
One of the long-term goals of artificial life research is to create autonomous, self-motivated, and intelligent animats. We study an intrinsic motivation system for behavioral self-exploration based on the maximization of the predictive information using the Stumpy robot, which is the first evaluation of the algorithm on a real robot. The control is organized in a closed-loop fashion with a reactive controller that is subject to fast synaptic dynamics. Even though the available sensors of the robot produce very noisy and peaky signals, the self-exploration algorithm was successful and various emerging behaviors were observed.read more
Citations
More filters
Journal ArticleDOI
Simulation-based internal models for safer robots
TL;DR: Inspired by the problem of how mobile robots could move quickly and safely through crowds of moving humans, this paper presents experimental results which compare the performance of the internal simulation-based controller with a purely reactive approach as a proof-of-concept study for the practical use of simulation- based internal models.
Posted Content
Emergent Real-World Robotic Skills via Unsupervised Off-Policy Reinforcement Learning
TL;DR: This paper shows that a recently proposed unsupervised skill discovery algorithm can be extended into an efficient off-policy method, making it suitable for performing unsuper supervised reinforcement learning in the real world, and provides substantial improvement in learning efficiency, making reward-free real-world training feasible.
Journal ArticleDOI
Value systems for developmental cognitive robotics
TL;DR: The extent to which existing value systems support attention focus, learning and prediction in an unsupervised setting is examined, as well as the types of robots and applications in which they are used.
Morphological Computation: The Body as a Computational Resource
TL;DR: A number of remarkable implications are discussed when real physical bodies are employed as computational resources when the physical body as a Computational Resource is considered.
Journal ArticleDOI
Expanding the Active Inference Landscape: More Intrinsic Motivations in the Perception-Action Loop.
TL;DR: This article reconstructs the active inference approach, locate the original formulation within, and show how alternative intrinsic motivations can be used while keeping many of the original features intact, and illustrates the connection to universal reinforcement learning by means of the formalism.
References
More filters
Journal ArticleDOI
Internal models in the cerebellum
TL;DR: This review will focus on the possibility that the cerebellum contains an internal model or models of the motor apparatus, and the necessity of such a model and the evidence, based on the ocular following response, that inverse models are found within the Cerebellar circuitry.
Journal ArticleDOI
Intrinsic Motivation Systems for Autonomous Mental Development
TL;DR: The mechanism of Intelligent Adaptive Curiosity is presented, an intrinsic motivation system which pushes a robot towards situations in which it maximizes its learning progress, thus permitting autonomous mental development.
Book
How the Body Shapes the Way We Think: A New View of Intelligence
Rolf Pfeifer,Josh C. Bongard +1 more
TL;DR: In How the Body Shapes the Way The authors Think, Rolf Pfeifer and Josh Bongard demonstrate that thought is not independent of the body but is tightly constrained, and at the same time enabled, by it.
Journal ArticleDOI
Resilient machines through continuous self-modeling.
TL;DR: A robot is described that can recover from change autonomously, through continuous self-modeling, and this concept may help develop more robust machines and shed light on self- modeling in animals.
Proceedings ArticleDOI
Curious model-building control systems
TL;DR: A novel curious model-building control system is described which actively tries to provoke situations for which it learned to expect to learn something about the environment, based on Watkins' Q-learning algorithm.