Learn More
Movement Primitives (MP) are a well-established approach for representing modular and re-usable robot movement generators. Many state-of-the-art robot learning successes are based MPs, due to their compact representation of the inherently continuous and high dimensional robot movements. A major goal in robot learning is to combine multiple MPs as building(More)
Efficient skill acquisition is crucial for creating versatile robots. One intuitive way to teach a robot new tricks is to demonstrate a task and enable the robot to imitate the demonstrated behavior. This approach is known as imitation learning. Classical methods of imitation learning, such as inverse reinforcement learning or behavioral cloning, suffer(More)
Supersensitivity of adenylyl cyclase after exposure to inhibitory agonists is a general means of cellular adaptation. We hypothesized that such "crosstalk" between muscarinic cholinergic agonists, beta 1-adrenoceptors, and adenylyl cyclase may be an important mechanism of cardiac adaptation to interventions that enhance vagal activity. We used primary(More)
OBJECTIVE The purpose of the present study was to investigate the comorbidity of personality disorders in patients with primary dysthymia compared to those with episodic major depression. METHOD A total of 177 out-patients with primary dysthymia and 187 outpatients with episodic major depression were administered a structured diagnostic interview for(More)
The acetylcholine receptor has been effectively solubilized from Torpedo californica electroplax under defined conditions with the nonionic detergent, beta-D-octylglucopyranoside. Preferential solubilization of the receptor protein, with regard to yield and specific alpha-bungarotoxin binding activity, occurs in the absence of salt and diminishes when NaCl(More)
Many Stochastic Optimal Control (SOC) approaches rely on samples to either obtain an estimate of the value function or a linearisation of the underlying system model. However, these approaches typically neglect the fact that the accuracy of the policy update depends on the closeness of the resulting trajectory distribution to these samples. The greedy(More)
One of the most elegant ways of teaching new skills to robots is to provide demonstrations of a task and let the robot imitate this behavior. Such imitation learning is a non-trivial task: Different anatomies of robot and teacher, and reduced robustness towards changes in the control task are two major difficulties in imitation learning. We present an(More)
A promising idea for scaling robot learning to more complex tasks is to use elemental behaviors as building blocks to compose more complex behavior. Ideally, such building blocks are used in combination with a learning algorithm that is able to learn to select, adapt, sequence and co-activate the building blocks. While there has been a lot of work on(More)