skip to main content

Title: Three Artificial Spintronic Leaky Integrate-and-Fire Neurons
Due to their non-volatility and intrinsic current integration capabilities, spintronic devices that rely on domain wall (DW) motion through a free ferromagnetic track have garnered significant interest in the field of neuromorphic computing. Although a number of such devices have already been proposed, they require the use of external circuitry to implement several important neuronal behaviors. As such, they are likely to result in either a decrease in energy efficiency, an increase in fabrication complexity, or even both. To resolve this issue, we have proposed three individual neurons that are capable of performing these functionalities without the use of any external circuitry. To implement leaking, the first neuron uses a dipolar coupling field, the second uses an anisotropy gradient, and the third uses shape variations of the DW track.
; ; ; ; ; ; ; ; ; ;
Award ID(s):
1910800 1910997
Publication Date:
Journal Name:
Sponsoring Org:
National Science Foundation
More Like this
  1. Spintronic devices, especially those based on motion of a domain wall (DW) through a ferromagnetic track, have received a significant amount of interest in the field of neuromorphic computing because of their non-volatility and intrinsic current integration capabilities. Many spintronic neurons using this technology have already been proposed, but they also require external circuitry or additional device layers to implement other important neuronal behaviors. Therefore, they result in an increase in fabrication complexity and/or energy consumption. In this work, we discuss three neurons that implement these functions without the use of additional circuitry or material layers.
  2. Advances in machine intelligence have sparked interest in hardware accelerators to implement these algorithms, yet embedded electronics have stringent power, area budgets, and speed requirements that may limit nonvolatile memory (NVM) integration. In this context, the development of fast nanomagnetic neural networks using minimal training data is attractive. Here, we extend an inference-only proposal using the intrinsic physics of domain-wall MTJ (DW-MTJ) neurons for online learning to implement fully unsupervised pattern recognition operation, using winner-take-all networks that contain either random or plastic synapses (weights). Meanwhile, a read-out layer trains in a supervised fashion. We find our proposed design can approach state-of-the-art success on the task relative to competing memristive neural network proposals, while eliminating much of the area and energy overhead that would typically be required to build the neuronal layers with CMOS devices.
  3. The challenge of developing an efficient artificial neuron is impeded by the use of external CMOS circuits to perform leaking and lateral inhibition. The proposed leaky integrate-and-fire neuron based on the three terminal magnetic tunnel junction (3T-MTJ) performs integration by pushing its domain wall (DW) with spin-transfer or spin-orbit torque. The leaking capability is achieved by pushing the neurons’ DWs in the direction opposite of integration using a stray field from a hard ferromagnet or a non-uniform energy landscape resulting from shape or anisotropy variation. Firing is performed by the MTJ stack. Finally, analog lateral inhibition is achieved by dipolar field repulsive coupling from each neuron. An integrating neuron thus pushes slower neighboring neurons’ DWs in the direction opposite of integration. Applying this lateral inhibition to a ten-neuron output layer within a neuromorphic crossbar structure enables the identification of handwritten digits with 94% accuracy.
  4. Amid increasing demands from students and the public, universities have recently reinvigorated their efforts to increase the number of faculty from underrepresented populations. Although a myriad of piecemeal programs targeting individual recruitment and development have been piloted at several institutions, overall growth in faculty diversity remains almost negligible and highly localized. To bring about genuine change, we hypothesize a consortia approach that links individuals to hiring opportunities within a state university system might be more effective. Here we present a case study describing the progress of the NSF-funded Alliances for Graduate Education and the Professoriate (AGEP) PROMISE Academy Alliance, a consortium within the University System of Maryland (USM) collaborating to develop, implement, self-study, evaluate, and disseminate a unique postdoc-to-faculty conversion model in the biomedical sciences. The initiative centers on diversifying faculty across five institutions in the USM, including teaching-focused institutions, comprehensive universities, research institutions, and professional schools. Components of this approach include (1) enhanced recruiting and hiring practices to attract outstanding postdoctoral scholars from underrepresented backgrounds, (2) multi-institutional networking and professional development, and (3) facilitated processes to transition (or “convert”) postdocs into tenure-track positions at their postdoctoral institution or another institution in the state system. This model is distinct frommore »more deficit-based approaches because it goes beyond focusing on building the individual’s skills to enter the professoriate. This program restructures the traditionally short-term nature of postdoctoral employment and incorporates a pathway to a tenure-track professorship at the same institution or within the same statewide system where the postdoc is trained. This multi-institutional model leverages collaboration and distinct institutional strengths to create cross-institutional support, advocacy, and policy. Importantly, it uses a decentralized financial structure that makes this approach distinctly replicable. Recognizing the immediate need for more collaborative approaches to diversify faculty and a lack of literature about such approaches, this case study describes the development of, and potential benefits of, a state university system, as well as the qualitative lessons learned from self-study, internal evaluation, external evaluation, and NSF site visits. The AGEP PROMISE Academy can serve as a model for replication at other university systems hoping to diversify their faculty.« less
  5. As network, I/O, accelerator, and NVM devices capable of a million operations per second make their way into data centers, the software stack managing such devices has been shifting from implementations within the operating system kernel to more specialized kernel-bypass approaches. While the in-kernel approach guarantees safety and provides resource multiplexing, it imposes too much overhead on microsecond-scale tasks. Kernel-bypass approaches improve throughput substantially but sacrifice safety and complicate resource management: if applications are mutually distrusting, then either each application must have exclusive access to its own device or else the device itself must implement resource management. This paper shows how to attain both safety and performance via intra-process isolation for data plane libraries. We propose protected libraries as a new OS abstraction which provides separate user-level protection domains for different services (e.g., network and in-memory database), with performance approaching that of unprotected kernel bypass. We also show how this new feature can be utilized to enable sharing of data plane libraries across distrusting applications. Our proposed solution uses Intel's memory protection keys (PKU) in a safe way to change the permissions associated with subsets of a single address space. In addition, it uses hardware watch-points to delay asynchronous eventmore »delivery and to guarantee independent failure of applications sharing a protected library. We show that our approach can efficiently protect high-throughput in-memory databases and user-space network stacks. Our implementation allows up to 2.3 million library entrances per second per core, outperforming both kernellevel protection and two alternative implementations that use system calls and Intel's VMFUNC switching of user-level address spaces, respectively.« less