Moving Sofa Problem
A geometric optimization problem in mathematics that seeks the largest-area shape able to pass around the corner of an L-shaped corridor of unit width; it was posed by Leo Moser in 1966.
Timeline (last 90 days)
Gerver's proof of the sofa's optimality has not yet been fully rigorously verified.
Meda modeled the moving sofa problem as a reinforcement learning environment, in which the agent controls the sofa's translation and rotation with the goal of maximizing the area that can pass through the corridor.
The moving sofa problem was first posed by Leo Moser in 1966: in a right-angled corridor of width 1, what is the maximum area of a sofa that can pass through the corner?
A developer without a deep mathematics background used the PPO algorithm to reconstruct an approximate shape of Gerver's sofa through reinforcement learning.
In this project, the agent models the problem as the task of controlling a sofa moving through a corridor; the sofa is defined by a series of points, and the agent learns to adjust the positions of these points so that the sofa can pass through the corner while maximizing its area.
All Facts (5)
The moving sofa problem was first posed by Leo Moser in 1966: in a right-angled corridor of width 1, what is the maximum area of a sofa that can pass through the corner?
60%UnverifiedGerver's proof of the sofa's optimality has not yet been fully rigorously verified.
50%UnverifiedMeda modeled the moving sofa problem as a reinforcement learning environment, in which the agent controls the sofa's translation and rotation with the goal of maximizing the area that can pass through the corridor.
50%UnverifiedIn this project, the agent models the problem as the task of controlling a sofa moving through a corridor; the sofa is defined by a series of points, and the agent learns to adjust the positions of these points so that the sofa can pass through the corner while maximizing its area.
50%UnverifiedA developer without a deep mathematics background used the PPO algorithm to reconstruct an approximate shape of Gerver's sofa through reinforcement learning.
50%