Abstract
Goal-driven deep learning produced significant advances in perception modelling. Models, however, often implement a single sensory domain and thus isolate a specific function. In this work, we go a step beyond and close the perception-action loop with a model of the frontoparietal network. The model implements biologically plausible macro-level structure by connecting cell count-fitted sensorimotor regions by pathways extracted from structural connectivity data. The model interfaces an anthropomorphic robotic hand and is trained to manipulate objects. We show that the biologically-inspired architecture significantly outperforms an architecture used in state-of-the-art robotics while converging substantially faster and relying only on raw sensory data. Moreover, preliminary in silico decoding analyses show promise in aligning with in vivo expectations.
Original language | English |
---|---|
Title of host publication | 2023 Conference on Cognitive Computational Neuroscience |
Publisher | Cognitive Computational Neuroscience |
Pages | 147-150 |
Number of pages | 4 |
DOIs | |
Publication status | Published - 24 Aug 2023 |
Event | Conference on Cognitive Computational Neuroscience 2023 - Oxford, United Kingdom Duration: 24 Aug 2023 → 27 Aug 2023 https://2023.ccneuro.org/index.php |
Conference
Conference | Conference on Cognitive Computational Neuroscience 2023 |
---|---|
Country/Territory | United Kingdom |
City | Oxford |
Period | 24/08/23 → 27/08/23 |
Internet address |