Organizers
- Vittorio Fra, Polytechnic of Turin, Italy
- Yulia Sandamirskaya, Zurich University of Applied Sciences, Switzerland
- Jens Egholm Pedersen, Technical University of Denmark, Denmark
- Chiara Bartolozzi, Foundation Italian Institute of Technology, Italy
- Giulia D’Angelo, Czech Technical University in Prague
- Claudio Gallicchio, University of Pisa, Italy
Abstract
Perception is the interface between sensing and knowledge, part of the process leading from stimulus acquisition to information handling. For animals, the nervous system processes signals from a distributed system of highly specialized peripheral sensors, collected into a single centralized brain.
Consequently, as long as neuromorphic engineering, encompassing neuromorphic computing and Neuro-AI, aims to emulate, or draw inspiration from, the nervous system, a fundamental aspect to be considered is the interplay between peripheral sensing and central processing. Essential for perception, this is also a key aspect of knowledge formation for intelligent systems.
Depending on the timescale required for action, perception triggers quick reaction or reasoning. Additionally, the different sensing modalities, i.e. tactile, visual or others, can define the actual interaction with the surrounding environment, leading to different ways of understanding or navigating the scene.
This Special Session targets perception through neuro-inspired approaches at both the sensory and the processing level, focusing on the formation of actionable knowledge through event-based sensing or computation.
Theoretical and application-oriented contributions are encouraged, in order to foster a comprehensive discussion on the results, strategies and prospects of bio-inspired perception.
List of Topics Covered in the Special Session
- Event-based sensing
- Signal encoding
- Spiking neural networks
- Analog computing
- Bio-inspired learning algorithms
- Spatial intelligence
- Autonomous navigation
- Neuromorphic Reservoir computing
- Bio-inspired hardware substrates
- Hardware-software co-design
- Cross-platform representation
- Embodied AI
- Profiling and benchmarking
- Edge computing
