Journal
ACS PHOTONICS
Volume 9, Issue 5, Pages 1647-1654Publisher
AMER CHEMICAL SOC
DOI: 10.1021/acsphotonics.1c02016
Keywords
resonance; temporal memory; recurrent neural networks; vowel classification; LSTM
Categories
Funding
- DARPA [FA8650-20-1-7028]
- DARPA NLM program [HR00111820046]
Ask authors/readers for more resources
There is a strong interest in using physical waves for artificial neural computing due to their fast speed and intrinsic parallelism. Resonance, a common feature in many wave systems, is a natural choice for analog computing in temporal signals. This study demonstrates that resonance can be utilized to construct stable and scalable recurrent neural networks. By incorporating resonators with different lifetimes, the computing system can develop both short-term and long-term memories simultaneously.
There is a strong interest in using physical waves for artificial neuralcomputing because of their unique advantages in fast speed and intrinsicparallelism. Resonance, as a ubiquitous feature across many wave systems, is anatural candidate for analog computing in temporal signals. We demonstrate thatresonance can be used to construct stable and scalable recurrent neural networks.By including resonators with different lifetimes, the computing system developsboth short-term and long-term memories simultaneously
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available