Abstract: In the study of embodied... In this paper, we evolve neural controllers for nine different simulated, legged agents. The agents have different body shapes, and differing numbers of legs. In this study we used one tripedal agent, four quadrupedal agents, two agents with five and seven points of contact with the ground plane, a salamandertype agent with nine points of contact, and two segmented agents with 10 points of contact. Despite the differing morphologies, each agent contains the same number of sensors and motors, and identical neural architectures. By randomizing the output values of the single hidden neurons from evolved neural controllers, it was found that for some agents, sensor-motor mappings are distributed evenly across the hidden layer, but that for other agents the distribution was less even. This trend was found to hold for evolved neural networks with hidden layers containing both three and five neurons (see Fig. 1). This suggests that particular morphological aspects (in this case, number of legs) have an effect on how sensor-motor mappings are distributed across a neural network when the weight space is evolved. This research is a first attempt to elucidate how evolved behaviours cause (or fail to cause) the centralization of neural structure. Fig 1. Effect of randomization of hidden neurons. The left-hand panel shows the fitness (distance travelled in meters) for each of the nine agents with the best evolved neural network (black bars). The white and gray bars indicate the distance travelled by the agent when each of the three hidden neurons, in turn, output random activations. The right-hand panel shows the performances for the nine agents using their best evolved neural networks, with five neurons in the hidden layer. The white and gray bars indicate the performances when each o...
**May not be in order
[edit database entry]
Bongard's work focuses on understanding the general nature of cognition, regardless of whether it is found in humans, animals or robots. This unique approach focuses on the role that morphology and evolution plays in cognition. Addressing these questions has taken him into the fields of biology, psychology, engineering and computer science.
Danforth is an applied mathematician interested in modeling a variety of physical, biological, and social phenomenon. He has applied principles of chaos theory to improve weather forecasts as a member of the Mathematics and Climate Research Network, and developed a real-time remote sensor of global happiness using messages from Twitter: the Hedonometer. Danforth co-runs the Computational Story Lab with Peter Dodds, and helps run UVM's reading group on complexity.
Laurent studies the interaction of structure and dynamics. His research involves network theory, statistical physics and nonlinear dynamics along with their applications in epidemiology, ecology, biology, and sociology. Recent projects include comparing complex networks of different nature, the coevolution of human behavior and infectious diseases, understanding the role of forest shape in determining stability of tropical forests, as well as the impact of echo chambers in political discussions.
Hines' work broadly focuses on finding ways to make electric energy more reliable, more affordable, with less environmental impact. Particular topics of interest include understanding the mechanisms by which small problems in the power grid become large blackouts, identifying and mitigating the stresses caused by large amounts of electric vehicle charging, and quantifying the impact of high penetrations of wind/solar on electricity systems.
Bagrow's interests include: Complex Networks (community detection, social modeling and human dynamics, statistical phenomena, graph similarity and isomorphism), Statistical Physics (non-equilibrium methods, phase transitions, percolation, interacting particle systems, spin glasses), and Optimization(glassy techniques such as simulated/quantum annealing, (non-gradient) minimization of noisy objective functions).