Event
Conditional Neural Processes for Self-Supervised Goal Discovery in Action Imitation
1 reports · 1 sources
Marco Gabriele Fedozzi and co-authors published a study evaluating whether Conditional Neural Processes can autonomously learn action representations that are informative about latent action goals. The authors report that, after adding an attention mechanism to the latent space, the model can form robust goal representations early in the observation of action sequences, and on this basis they suggest the model has potential as a general-purpose self-supervised learner for robots. The work was presented as a poster at the IROS2023 Workshop on World Models and Predictive Coding in Cognitive Robotics; the above effects are the authors' claims in the paper.
Report timeline
- Conditional Neural Processes for Self-Supervised Goal Discovery in Action Imitation
The study evaluates whether Conditional Neural Processes can autonomously learn action representations that are informative about underlying goals. It finds that adding an attention mechanism to the latent space enables robust goal representations to form early in the observation of an action sequence. The authors accordingly suggest the model could serve as a general self-supervised learner for robots. The work was presented as a poster at the IROS 2023 Workshop on World Models and Predictive Coding in Cognitive Robotics.
arXiv cs.RO ·