Search Results for author:
Found 12 papers, 6 papers with code
Geodesics, Non-linearities and the Archive of Novelty Search
An argument that is often encountered in the literature is that the archive prevents exploration from backtracking or cycling, i. e. from revisiting previously encountered areas in the behavior space.
Few-shot Quality-Diversity Optimization
Experiments carried in both sparse and dense reward settings using robotic manipulation and navigation benchmarks show that it considerably reduces the number of generations that are required for QD optimization in these environments.
BR-NS: an Archive-less Approach to Novelty Search
In this paper, we discuss an alternative approach to novelty estimation, dubbed Behavior Recognition based Novelty Search (BR-NS), which does not require an archive, makes no assumption on the metrics that can be defined in the behavior space and does not rely on nearest neighbours search.
Sparse Reward Exploration via Novelty Search and Emitters
Contrary to existing emitters-based approaches, SERENE separates the search space exploration and reward exploitation into two alternating processes.
Novelty Search makes Evolvability Inevitable
Evolvability is thus a natural byproduct of the search in this context.
DREAM Architecture: a Developmental Approach to Open-Ended Learning in Robotics
Robots are still limited to controlled conditions, that the robot designer knows with enough details to endow the robot with the appropriate models or behaviors.
State Representation Learning from Demonstration
Indeed, a compact representation of such a state is beneficial to help robots grasp onto their environment for interacting.
Unsupervised Learning and Exploration of Reachable Outcome Space
Results show that TAXONS can find a diverse set of controllers, covering a good part of the ground-truth outcome space, while having no information about such space.
Building an Affordances Map with Interactive Perception
Robots need to understand their environment to perform their task.
Bootstrapping Robotic Ecological Perception from a Limited Set of Hypotheses Through Interactive Perception
A robot with the ability to build and adapt this interpretation process according to its own tasks and capabilities would push away the limits of what robots can achieve in a non controlled environment.
From exploration to control: learning object manipulation skills through novelty search and local adaptation
The approach has been validated on two different experiments on the Baxter robot: a ball launching and a joystick manipulation tasks.
Quick and energy-efficient Bayesian computing of binocular disparity using stochastic digital signals
Reconstruction of the tridimensional geometry of a visual scene using the binocular disparity information is an important issue in computer vision and mobile robotics, which can be formulated as a Bayesian inference problem.
