IC Colloquium: Neuronal Architectures for Embodied AI and Cognition
By: Yulia Sandamirskaya - Intel
IC/SV Faculty candidate
Abstract
Biological neural systems show a large variety of capabilities that we would like to achieve in technological systems. These capabilities include the ability to learn on the fly (sometimes, literally), to learn new tasks, and to adapt to a large variety of environmental conditions. Organism takle the complexity of their environment with a speed and efficiency as yet unreached in our digital solutions. Biological neural systems make use of a variety of computational units, circuit motives and network structures, dynamics, and learning processes that work at different spatial and temporal scales. Over the last few decades, many of these units and processes have been modeled theoretically and in simulation, leading us to understand their basic functioning. In this talk, I will review important computational principles that have been derived from the study of biological neural systems. Around one or two exemplary cases from my work, I will show how these principles have guided the development of functional neuronal architectures capable of generating behavior and of learning in closed behavioral loop, in some cases, literally on the fly.
Bio
Dr. Yulia Sandamirskaya leads the Applications Research team of the Neuromorphic Computing Lab at Intel. Her team develops spiking neuronal network based algorithms for neuromorphic hardware to demonstrate the potential of neuromorphic computing in real-world applications. She has 15 years of research experience in the fields of neural dynamics, embodied cognition, and autonomous robotics. She led a research group “Neuromorphic Cognitive Robots” at the Institute of Neuroinformatics of the University of Zurich and ETH Zurich, Switzerland and the “Autonomous learning” group at the Institute for Neural Computation at the Ruhr-University Bochum (RUB), Germany. She has coordinated the EU H2020 project NEUROTECH that aimed to create and support the neuromorphic computing community in Europe and co-chaired EUCog -- the association for artificial cognitive systems.
More information
IC/SV Faculty candidate
Abstract
Biological neural systems show a large variety of capabilities that we would like to achieve in technological systems. These capabilities include the ability to learn on the fly (sometimes, literally), to learn new tasks, and to adapt to a large variety of environmental conditions. Organism takle the complexity of their environment with a speed and efficiency as yet unreached in our digital solutions. Biological neural systems make use of a variety of computational units, circuit motives and network structures, dynamics, and learning processes that work at different spatial and temporal scales. Over the last few decades, many of these units and processes have been modeled theoretically and in simulation, leading us to understand their basic functioning. In this talk, I will review important computational principles that have been derived from the study of biological neural systems. Around one or two exemplary cases from my work, I will show how these principles have guided the development of functional neuronal architectures capable of generating behavior and of learning in closed behavioral loop, in some cases, literally on the fly.
Bio
Dr. Yulia Sandamirskaya leads the Applications Research team of the Neuromorphic Computing Lab at Intel. Her team develops spiking neuronal network based algorithms for neuromorphic hardware to demonstrate the potential of neuromorphic computing in real-world applications. She has 15 years of research experience in the fields of neural dynamics, embodied cognition, and autonomous robotics. She led a research group “Neuromorphic Cognitive Robots” at the Institute of Neuroinformatics of the University of Zurich and ETH Zurich, Switzerland and the “Autonomous learning” group at the Institute for Neural Computation at the Ruhr-University Bochum (RUB), Germany. She has coordinated the EU H2020 project NEUROTECH that aimed to create and support the neuromorphic computing community in Europe and co-chaired EUCog -- the association for artificial cognitive systems.
More information
Practical information
- General public
- Free
Contact
- Host: Paolo Ienne