Abstract
| Original language | Undefined |
|---|---|
| Title of host publication | The Eighth International Conference on Future Computational Technologies and Applications, FUTURE COMPUTING 2016 |
| Place of Publication | Wilmington, U.S.A. |
| Publisher | International Academy, Research and Industry Association |
| Pages | 15-20 |
| Number of pages | 6 |
| ISBN (Print) | 978-1-61208-461-9 |
| Publication status | Published - 20 Mar 2016 |
| Event | The Eighth International Conference on Future Computational Technologies and Applications, FUTURE COMPUTING 2016 - Rome, Italy Duration: 20 Mar 2016 → 24 Mar 2016 Conference number: 8 https://www.iaria.org/conferences2016/AwardsFUTURECOMPUTING16.html |
Publication series
| Name | |
|---|---|
| Publisher | IARIA International Academy, Research and Industry Association |
Conference
| Conference | The Eighth International Conference on Future Computational Technologies and Applications, FUTURE COMPUTING 2016 |
|---|---|
| Abbreviated title | FUTURE COMPUTING 2016 |
| Country | Italy |
| City | Rome |
| Period | 20/03/16 → 24/03/16 |
| Internet address |
Keywords
- MSC-65Z05
- EWI-26922
- Simulation
- Unconventional computation
- METIS-316877
- Nanoparticle network
- IR-100187
- Evolution-in-nanomaterio
- Neural network
Cite this
}
Unconventional computing using evolution-in-nanomaterio: neural networks meet nanoparticle networks. / Greff, Klaus; van Damme, Rudolf M.J.; Koutnik, Jan; Broersma, Haitze J.; Mikhal, Julia Olegivna; Lawrence, Celestine Preetham; van der Wiel, Wilfred Gerard; Schmidhuber, Jürgen.
The Eighth International Conference on Future Computational Technologies and Applications, FUTURE COMPUTING 2016. Wilmington, U.S.A. : International Academy, Research and Industry Association, 2016. p. 15-20.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review
TY - GEN
T1 - Unconventional computing using evolution-in-nanomaterio: neural networks meet nanoparticle networks
AU - Greff, Klaus
AU - van Damme, Rudolf M.J.
AU - Koutnik, Jan
AU - Broersma, Haitze J.
AU - Mikhal, Julia Olegivna
AU - Lawrence, Celestine Preetham
AU - van der Wiel, Wilfred Gerard
AU - Schmidhuber, Jürgen
N1 - eemcs-eprint-26922
PY - 2016/3/20
Y1 - 2016/3/20
N2 - Recently published experimental work on evolution-in-materio applied to nanoscale materials shows promising results for future reconfigurable devices. These experiments were performed on disordered nano-particle networks that have no predefined design. The material has been treated as a blackbox, and genetic algorithms have been used to find appropriate configuration voltages to enable the target functionality. In order to support future experiments, we developed simulation tools for predicting candidate functionalities. One of these tools is based on a physical model, but the one we introduce in this paper is based on an artificial neural network. The advantage of this newly presented approach is that, after training the neural network to match either the real material or its physical model, it can be configured using gradient descent instead of a black-box optimisation. The experiments we report here demonstrate that the neural network can model the simulated nano-material quite accurately. The differentiable, neural network-based material model is then used to find logic gates, as a proof of principle. This shows that the new approach has great potential for partly replacing costly and time-consuming experiments with the real materials. Therefore, this approach has a high relevance for future computing, either as an alternative to digital computing or as an alternative way of producing multi-functional reconfigurable devices.
AB - Recently published experimental work on evolution-in-materio applied to nanoscale materials shows promising results for future reconfigurable devices. These experiments were performed on disordered nano-particle networks that have no predefined design. The material has been treated as a blackbox, and genetic algorithms have been used to find appropriate configuration voltages to enable the target functionality. In order to support future experiments, we developed simulation tools for predicting candidate functionalities. One of these tools is based on a physical model, but the one we introduce in this paper is based on an artificial neural network. The advantage of this newly presented approach is that, after training the neural network to match either the real material or its physical model, it can be configured using gradient descent instead of a black-box optimisation. The experiments we report here demonstrate that the neural network can model the simulated nano-material quite accurately. The differentiable, neural network-based material model is then used to find logic gates, as a proof of principle. This shows that the new approach has great potential for partly replacing costly and time-consuming experiments with the real materials. Therefore, this approach has a high relevance for future computing, either as an alternative to digital computing or as an alternative way of producing multi-functional reconfigurable devices.
KW - MSC-65Z05
KW - EWI-26922
KW - Simulation
KW - Unconventional computation
KW - METIS-316877
KW - Nanoparticle network
KW - IR-100187
KW - Evolution-in-nanomaterio
KW - Neural network
M3 - Conference contribution
SN - 978-1-61208-461-9
SP - 15
EP - 20
BT - The Eighth International Conference on Future Computational Technologies and Applications, FUTURE COMPUTING 2016
PB - International Academy, Research and Industry Association
CY - Wilmington, U.S.A.
ER -