Evolutionary Robotics and the Radical Envelope-of-Noise Hypothesis

By N. Jakobi. 1997

How researchers see this paper

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

In Robotics & Control

Citations in the corpus

Domain randomization for transferring deep neural networks from simulation to the real world (2017, background): “In [18], the authors suggested that transferability can be achieved by randomly varying the items in the implementation set – model parameters that are not essential to the controller achieving near-optimal performance.”
Domain Randomization and Generative Models for Robotic Grasping (2017, background): “all aspects of the simulator that do have a basis in reality [13].”
The Transferability Approach: Crossing the Reality Gap in Evolutionary Robotics (2013, methodology): “In the current paper, the approach is additionally applied to a navigation task in a T-maze [26] and both applications allow systematic comparisons between the Transferability approach and state-of-the-art methods.”
Exploring the T-Maze: Evolving Learning-Like Robot Behaviors Using CTRNNs (2003, background): “Previous work on TMaze navigation in evolutionary robotics includes delayed response tasks where the robots had to perform one or several turns in a maze on the basis of light source cues given to the robot [5][13].”
Better Living Through Chemistry: Evolving GasNets for Robot Control (1998, methodology): “This methodology is described in detail elsewhere (Jakobi, 1998a, 1998b), but since the experiments reported in this paper make extensive use of it, we o er a brief sketch here: 1.”
Adversarial Environment Generation for Learning to Navigate the Web (2021, methodology): “Another option would be to apply domain randomization (DR) (as in e.g. Jakobi (1997); Sadeghi & Levine (2016); Tobin et al. (2017)) to randomize parameters of websites, or automatically increase some parameter controlling the difﬁculty over time (as in Gur et al. (2019)).”
A scalable pipeline for designing reconfigurable organisms (2020, background): “Previous work has shown that noise resistance in simulation is a simple and effective predictor of whether a design will maintain its behavior when instantiated physically (18).”
The evolvability of artificial neural net-works for robot control (2002, methodology): “9(a)) are required to navigate along a straight corridor, before turning either left or right at a T-junction (Jakobi, 1998a).”
Neuronal Plasticity and Temporal Adaptivity: GasNet Robot Control Networks (2002, methodology): “The level of noise in the model is higher than that found in the real world, to evolve robot controllers successfully in simulation able to operate successfully in the real world (Jakobi, 1998).”
Embodied Evolution: Distributing an evolutionary algorithm in a population of robots (2002, background): “Finally, Section 8 outlines some caveats regarding our approach, followed by conclusions in Section 9.”
The Player/Stage Project: Tools for Multi-Robot and Distributed Sensor Systems (2003, background): “Low fidelity simulation can actually be an advantage when designing robot controllers that must run on real robots, as it encourages the use of robust control techniques [13].”
Emergent Complexity and Zero-shot Transfer via Unsupervised Environment Design (2020, methodology): “Domain randomization (DR) [17] is an alternative approach in which a designer specifies a set of parameters to which the policy should be robust [32, 41, 40].”
Neuromodulation of Reactive Sensorimotor Mappings as a Short-Term Memory Mechanism in Delayed Response Tasks (2002, background): “Delayed response tasks, which are common in experimental psychology research, have recently also received much attention in the adaptive behavior and artificial neural network (ANN) research community (e.g. Ulbricht, 1996; Jakobi, 1997; Rylatt & Czarnecki, 1998, 2000; Linåker & Jacobsson, 2001, 2002”
Spike-Timing Dependent Plasticity for Evolved Robots (2002, background): “Not for commercial use or unauthorized distribution. at PENNSYLVANIA STATE UNIV on April 10, 2008 http://adb.sagepub.comDownloaded from imal simulation design strategy (Jakobi, 1997).”