Learning
Deliverable No
Deliverable title
Delivery date
D6.9
Publication of RL on ARMAR: Report supplemented by one or two publications on RL for glass-filling and on the results of applying RL to high-dimensional action spaces.
D6.10
Scientific publication on a general decision making framework which integrates a learner, a planner and a human teacher to learn and execute complex tasks in human scenarios.
D6.8
Scientific Publication on the analytical solution for the weight growth in time-continuous spike-timing-dependent plasticity.
D6.7
Publication of neuronally motivated disturbance compensation learning methods on a multi-joint arm.
D6.6
On-line Learning of Macro Planning Operators using Probabilistic Estimations of Cause-Effects
D6.5
Technical Report/Scientific publication on "Implementation of nSARSA with function approximation in its two stages as given by tasks 6.8.1 and 6.8.2 first on a simple platform and then on ARMAR".
D6.4
Technical Report/Scientific publication Theoretical concepts of contingency minimization and return maximization in a conjoint way for combining CL with RL.
Strategies for modification of Actor-Critic loops
D6.2
Chaining learning architectures in a simple closed-loop behavioural context (scientific publication submitted to Biological Cybernetics)
D6.1
Neuronal Network Simulator
Publication of RL on ARMAR: Report supplemented by one or two publications on RL for glass-filling and on the results of applying RL to high-dimensional action spaces.