|
Open-ended learning in robotics| old_uid | 16232 |
|---|
| title | Open-ended learning in robotics |
|---|
| start_date | 2018/09/05 |
|---|
| schedule | 12h15 |
|---|
| online | no |
|---|
| location_info | Amphi Herbrand |
|---|
| details | séminaire axe neuro |
|---|
| summary | Reinforcement learning is a reward-driven adaptation process. This formalism has been very successful in defining learning processes in both robotics and neuroscience. However, it requires the definition of a state space and an action space that the learning will explore. These state and action spaces determine the type of task that the robot can perform. Moreover this definition requires a significant expertise and the performance of the learning process is strongly dependent on this choice. The remarkable human cognitive abilities are, according to A. Karmiloff-Smith, related to our capacity to rewrite our knowledge in a different format, potentially more suited to solving the tasks we face. The European project DREAM aims to provide robots with such a capacity by focusing on the construction and redescription of state and action spaces. The project focuses in particular on the very first steps of this process, the creation of state spaces and actions starting from very limited knowledge about the environment and its characteristics. The results obtained will be presented as well as the links of this work with neuroscience. |
|---|
| responsibles | <not specified> |
|---|
| |
|