49 citations · 131 across the 37 of their papers we have counts for
8 papers · 1 filter
Deep-learning-based identification of individual motion characteristics from upper-limb trajectories towards disorder stage evaluation
Tim Sziburis, Susanne Blex, Tobias Glasmachers +1
The identification of individual movement characteristics sets the foundation for the assessment of personal rehabilitation progress and can provide diagnostic information on level…
The (1+1)-ES Reliably Overcomes Saddle Points
Tobias Glasmachers
It is known that step size adaptive evolution strategies (ES) do not converge (prematurely) to regular points of continuously differentiable objective functions. Among critical poi…
Challenges of Convex Quadratic Bi-objective Benchmark Problems
Tobias Glasmachers
Convex quadratic objective functions are an important base case in state-of-the-art benchmark collections for single-objective optimization on continuous domains. Although often co…
Challenges in High-dimensional Reinforcement Learning with Evolution Strategies
Nils Müller, Tobias Glasmachers
Evolution Strategies (ESs) have recently become popular for training deep neural networks, in particular on reinforcement learning tasks, a special form of controller design. Compa…
Drift Theory in Continuous Search Spaces: Expected Hitting Time of the (1+1)-ES with 1/5 Success Rule
Youhei Akimoto, Anne Auger, Tobias Glasmachers
This paper explores the use of the standard approach for proving runtime bounds in discrete domains---often referred to as drift analysis---in the context of optimization on a cont…
Global Convergence of the (1+1) Evolution Strategy
Tobias Glasmachers
We establish global convergence of the (1+1) evolution strategy, i.e., convergence to a critical point independent of the initial state. More precisely, we show the existence of a…