Abstract
Evaluating the effectiveness and performance of neuromorphic hardware is difficult. It is even more difficult when the task of interest is a closed-loop task; that is, a task where the output from the neuromorphic hardware affects some environment, which then in turn affects the hardware's future input. However, closed-loop situations are one of the primary potential uses of neuromorphic hardware. To address this, we present a methodology for generating closed-loop benchmarks that makes use of a hybrid of real physical embodiment and a type of “minimal” simulation. Minimal simulation has been shown to lead to robust real-world performance, while still maintaining the practical advantages of simulation, such as making it easy for the same benchmark to be used by many researchers. This method is flexible enough to allow researchers to explicitly modify the benchmarks to identify specific task domains where particular hardware excels. To demonstrate the method, we present a set of novel benchmarks that focus on motor control for an arbitrary system with unknown external forces. Using these benchmarks, we show that an error-driven learning rule can consistently improve motor control performance across a randomly generated family of closed-loop simulations, even when there are up to 15 interacting joints to be controlled.
Highlights
Neuromorphic hardware holds great promise for a wide variety of applications
As we argued here, benchmarking against a wide variety of randomly generated minimal simulation systems is needed to demonstrate the space of potential situations in which neuromorphic adaptive control performs well
While the particular minimal simulation shown here suggests that this adaptive control algorithm is worth further investigation, the overall goal of this paper is to present the general idea of using minimal simulation as a way to benchmark neuromorphic hardware
Summary
Neuromorphic hardware holds great promise for a wide variety of applications. The combination of massively parallel computation and low power consumption means that there is the potential to have complex algorithms running in embedded processing situations, without being a significant drain on available energy. A crucial challenge is to identify what sort of always-on or interactive functionality can best exploit these devices. To evaluate applications of neuromorphic hardware, we need benchmark tasks. These tasks must allow us to compare across different instances of neuromorphic hardware (and potentially across different algorithms implemented in that hardware). Good benchmarks will allow us to quantitatively compare systems, letting researchers both measure the progress in the field, and directly compare competing approaches
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
