J. Parker Mitchell

J. Parker Mitchell

Auto-generated profile, based on publicly available information

J. Parker Mitchell

Oak Ridge National Laboratory

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 October 2017
Verified
Conference Paper
Closed access
JMJ. Parker Mitchell
J. Parker Mitchell
GBGrant Bruer
Grant Bruer
MDMark E. Dean
Mark E. Dean
JPJames S. Plank
James S. Plank
GRGarrett Rose
Garrett Rose
CSCatherine D. Schuman
Catherine D. Schuman
In this paper we describe the use of a new neuromorphic computing framework to implement the navigation system for a roaming, obstacle avoidance robot. Using a Dynamic Adaptive Neural Network Array (DANNA) structure, our TENNLab (Laboratory of Tennesseans Exploring Neural Networks) hardware/software co-design framework and evolutionary optimization (EO) as the training algorithm, we create, train, implement, and test a spiking neural network autonomous robot control system using an array of neuromorphic computing elements built on an FPGA. The simplicity and flexibility of the DANNA neuromorphic computing elements allow for sufficient scale and connectivity on a Xilinx Kintex-7 FPGA to support sensory input and motor control for a mobile robot to navigate a dynamically changing environment. We further describe how more complex capabilities can be added using the same platform, e.g. object identification and tracking.

27 July 2021
Verified
Conference Paper
Open access
RPRobert Patton
Robert Patton
CSCatherine Schuman
Catherine Schuman
SKShruti Kulkarni
Shruti Kulkarni
MPMaryam Parsa
Maryam Parsa
JMJ. Parker Mitchell
J. Parker Mitchell
SSShay Snyder
Shay Snyder
Neuromorphic computing has many opportunities in future autonomous systems, especially those that will operate at the edge. However, there are relatively few demonstrations of neuromorphic implementations on real-world applications, partly because of the lack of availability of neuromorphic hardware and software, but also because of the lack of availability of an accessible demonstration platform. In this work, we propose utilizing the F1Tenth platform as an evaluation task for neuromorphic computing. F1Tenth is a competition wherein one tenth scale cars compete in an autonomous racing task; there are significant open source resources in both software and hardware for realizing this task. We present a workflow with neuromorphic hardware, software, and training that can be used to develop a spiking neural network for neuromorphic hardware deployment to perform autonomous racing. We present initial results on utilizing this approach for this small-scale, real-world autonomous vehicle task.

23 July 2018
Verified
Conference Paper
Open access
PDF file
JMJ. Parker Mitchell
J. Parker Mitchell
MDMark E. Dean
Mark E. Dean
GBGrant R. Bruer
Grant R. Bruer
JPJames S. Plank
James S. Plank
GRGarrett Rose
Garrett Rose

1 December 2019
Verified
Conference Paper
Closed access
MPMaryam Parsa
Maryam Parsa
JMJ. Parker Mitchell
J. Parker Mitchell
CSCatherine D. Schuman
Catherine D. Schuman
RPRobert M. Patton
Robert M. Patton
TPThomas E. Potok
Thomas E. Potok
KRKaushik Roy
Kaushik Roy
Designing a neuromorphic computing system involves selection of several hyperparameters that not only affect the accuracy of the framework, but also the energy efficiency and speed of inference and training. These hyperparameters might be inherent to the training of the spiking neural network (SNN), the input/output encoding of the real-world data to spikes, or the underlying neuromorphic hardware. In this work, we present a Bayesian-based hyperparameter optimization approach for spiking neuromorphic systems, and we show how this optimization framework can lead to significant improvement in designing accurate neuromorphic computing systems. In particular, we show that this hyperparameter optimization approach can discover the same optimal hyperparameter set for input encoding as a grid search, but with far fewer evaluations and far less time. We also show the impact of hardware-specific hyperparameters on the performance of the system, and we demonstrate that by optimizing these hyperparameters, we can achieve significantly better application performance.

1 November 2016
Verified
Conference Paper
Closed access
CSCatherine Schuman
Catherine Schuman
ADAdam Disney
Adam Disney
SSSusheela P. Singh
Susheela P. Singh
GBGrant Bruer
Grant Bruer
JMJ. Parker Mitchell
J. Parker Mitchell
JPJames S. Plank
James S. Plank
One of the key impediments to the success of current neuromorphic computing architectures is the issue of how best to program them. Evolutionary optimization (EO) is one promising programming technique; in particular, its wide applicability makes it especially attractive for neuromorphic architectures, which can have many different characteristics. In this paper, we explore different facets of EO on a spiking neuromorphic computing model called DANNA. We focus on the performance of EO in the design of our DANNA simulator, and on how to structure EO on both multicore and massively parallel computing systems. We evaluate how our parallel methods impact the performance of EO on Titan, the U.S.'s largest open science supercomputer, and BOB, a Beowulf-style cluster of Raspberry Pi's. We also focus on how to improve the EO by evaluating commonality in higher performing neural networks, and present the result of a study that evaluates the EO performed by Titan.

1 July 2020
Verified
Conference Paper
Open access
MPMaryam Parsa
Maryam Parsa
CSCatherine Schuman
Catherine Schuman
PDPrasanna Date
Prasanna Date
DRDerek C. Rose
Derek C. Rose
BKBill Kay
Bill Kay
KRKaushik Roy
Kaushik Roy
Training neural networks for neuromorphic deployment is non-trivial. There have been a variety of approaches proposed to adapt back-propagation or back-propagation-like algorithms appropriate for training. Considering that these networks often have very different performance characteristics than traditional neural networks, it is often unclear how to set either the network topology or the hyperparameters to achieve optimal performance. In this work, we introduce a Bayesian approach for optimizing the hyperparameters of an algorithm for training binary communication networks that can be deployed to neuromorphic hardware. We show that by optimizing the hyperparameters on this algorithm for each dataset, we can achieve improvements in accuracy over the previous state-of-the-art for this algorithm on each dataset (by up to 15 percent). This jump in performance continues to emphasize the potential when converting traditional neural networks to binary communication applicable to neuromorphic hardware.