Ryan Dellana




Google Scholar

(505) 284-3678

Sandia National Laboratories, New Mexico
P.O. Box 5800
Albuquerque, NM 87185-1327


Dr. Ryan Dellana is a Postdoc in the department of Cognitive and Emerging Computing (1421). His primary interest is in developing Cognitive Architectures for robots. He is the creator and maintainer of the Cognitive Architecture Working Group (wg-CAWG), connecting people across all of SNL who share this interest. He designs/constructs both simulated and physical robots to serve as embodiments to test brain-inspired agents and neuromorphic algorithms. His skills include Machine Learning, Computer Vision, Robotics, and Edge Computing.

Image of blathers
Event Camera Turret designed for Neural Engineering Research Lab (NERL)


Doctor of Philosophy in Computer Science – North Carolina Agricultural and Technical State University — 2018

Master of Science in Computer Science – East Carolina University — 2015

Bachelor of Science in Computer Science – East Carolina University — 2009


Severa, William, Craig M. Vineyard, Ryan Dellana, Stephen J. Verzi, and James B. Aimone. Training deep neural networks for binary communication with the whetstone method. Nature Machine Intelligence 1, no. 2 (2019): 86-94.

Vineyard, Craig M., Ryan Dellana, James B. Aimone, Fredrick Rothganger, and William M. Severa. Low-power deep learning inference using the SpiNNaker neuromorphic platform. In Proceedings of the 7th Annual Neuro-inspired Computational Elements Workshop (NICE), pp. 1-7. 2019.

Bennett CH, Xiao TP, Dellana R, Feinberg B, Agarwal S, Marinella MJ, Agrawal V, Prabhakar V, Ramkumar K, Hinh L, Saha S. Device-aware inference operations in SONOS nonvolatile memory arrays. In 2020 IEEE International Reliability Physics Symposium (IRPS) 2020 Apr 28 (pp. 1-6). IEEE.

*Bennett CH, *Dellana R, Xiao TP, Feinberg B, Agarwal S, Cardwell S, Marinella MJ, Severa W, Aimone B. (*joint first authors) Evaluating complexity and resilience trade-offs in emerging memory inference machines. In Proceedings of the Neuro-inspired Computational Elements Workshop 2020 Mar 17 (pp. 1-4).

Parsa M, Schuman CD, Date P, Rose DC, Kay B, Mitchell JP, Young SR, Dellana R, Severa W, Potok TE, Roy K. Hyperparameter optimization in binary communication networks for neuromorphic deployment. In 2020 International Joint Conference on Neural Networks (IJCNN) 2020 Jul 19 (pp. 1-9).

Severa W, Dellana R, Vineyard CM. Effective pruning of binary activation neural networks. In International Conference on Neuromorphic Systems 2020 2020 Jul 28 (pp. 1-5).

Ryan Dellana, Felix Wang, William M. Severa, Esteban J. Guillen, Jaimie Miller Murdock. MIKE – Multi-task Implicit Knowledge Embeddings by Autoencoding through a Shared Input Space. (Submitted ICLR 2022)

Vineyard, Craig M., William M. Severa, Aaron Hill, James B. Aimone, Esteban J. Guillen, Javier Zazueta, and Ryan Dellana. "Spiking neural approaches for SAR ATR." In Automatic Target Recognition XXXII, vol. 12096. SPIE, 2022.

Patents & Trademarks

W. Severa, R. Dellana, C. Vineyard, J. Aimone “SYSTEM AND METHOD FOR TRAINING DEEP ARTIFICIAL NEURAL NETWORKS” (SD14643.0 Non-Provisional Application Serial No. 16/146,904 filed 9/28/2018)

A. Anwar, R. Dellana, W. Severa, C. Vineyard “NEURAL NETWORK ROBUSTNESS VIA BINARY ACTIVATION” (SD15133.0 Non-Provisional Application Serial No. 17/314,751 filed 5/07/2021)