Kenneth Stewart

Kenneth Stewart

Auto-generated profile, based on publicly available information

Kenneth Stewart

Accenture Labs, United States

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
19 October 2020 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
Verified
Journal Article
Open access
KSKenneth Stewart
Kenneth Stewart
GOGarrick Orchard
Garrick Orchard
SSSumit Bam Shrestha
Sumit Bam Shrestha
ENEmre Neftci
Emre Neftci
We present the Surrogate-gradient Online Error-triggered Learning (SOEL) system for online few-shot learning on neuromorphic processors. The SOEL learning system uses a combination of transfer learning and principles of computational neuroscience and deep learning. We show that partially trained deep s (s) implemented on neuromorphic hardware can rapidly adapt online to new classes of data within a domain. SOEL updates trigger when an error occurs, enabling faster learning with fewer updates. Using gesture recognition as a case study, we show SOEL can be used for online few-shot learning of new classes of pre-recorded gesture data and rapid online learning of new gestures from data streamed live from a Dynamic Active-pixel Vision Sensor to an Intel Loihi neuromorphic research processor.

30 September 2022 in Neuromorphic Computing and Engineering
Verified
Journal Article
Open access
PDF file
KSKenneth Michael Stewart
Kenneth Michael Stewart
ENEmre O Neftci
Emre O Neftci
Adaptive ‘life-long’ learning at the edge and during online task performance is an aspirational goal of artificial intelligence research. Neuromorphic hardware implementing spiking neural networks (SNNs) are particularly attractive in this regard, as their real-time, event-based, local computing paradigm makes them suitable for edge implementations and fast learning. However, the long and iterative learning that characterizes state-of-the-art SNN training is incompatible with the physical nature and real-time operation of neuromorphic hardware. Bi-level learning, such as meta-learning is increasingly used in deep learning to overcome these limitations. In this work, we demonstrate gradient-based meta-learning in SNNs using the surrogate gradient method that approximates the spiking threshold function for gradient estimations. Because surrogate gradients can be made twice differentiable, well-established, and effective second-order gradient meta-learning methods such as model agnostic meta learning (MAML) can be used. We show that SNNs meta-trained using MAML perform comparably to conventional artificial neural networks meta-trained with MAML on event-based meta-datasets. Furthermore, we demonstrate the specific advantages that accrue from meta-learning: fast learning without the requirement of high precision weights or gradients, training-to-learn with quantization and mitigating the effects of approximate synaptic plasticity rules. Our results emphasize how meta-learning techniques can become instrumental for deploying neuromorphic learning technologies on real-world problems.

19 April 2023
Verified
Conference Paper
Open access
ADAndreea Danielescu
Andreea Danielescu
SHSharone A Horowit-Hendler
Sharone A Horowit-Hendler
APAlexandria Pabst
Alexandria Pabst
KSKenneth Michael Stewart
Kenneth Michael Stewart
EGEric M Gallo
Eric M Gallo
MAMatthew Peter Aylett
Matthew Peter Aylett
As voice assistant usage continues to grow, their homogeneity becomes even more problematic with the UNESCO report, “I’d Blush if I could” showing that designing only feminine voice assistants encourages negative behavior, both with virtual assistants and with real people [3]. While masculine text-to-speech (TTS) voices exist, ones that cover the full range of gender presentations, such as non-binary or gender-ambiguous voices are largely missing. In this paper, we present a method of creating a non-binary TTS voice and an example voice, Sam, created with input from the non-binary and transgender communities. We have open-sourced the resulting voice, along with the process and data used to create it. Finally, we present results from a large-scale survey showing that non-binary individuals are more likely to prefer a non-binary voice assistant compared to cisgendered individuals and discuss differences across age and gender.

11 April 2023
Verified
Conference Paper
Open access
KSKenneth Michael Stewart
Kenneth Michael Stewart
TSTimothy Shea
Timothy Shea
NPNoah Pacik-Nelson
Noah Pacik-Nelson
EGEric Gallo
Eric Gallo
ADAndreea Danielescu
Andreea Danielescu
Despite the maturity and availability of speech recognition systems, there are few available spiking speech recognition tasks that can be implemented with current neuromorphic systems. The methods used previously to generate spiking speech data are not capable of encoding speech in real-time or encoding very large modern speech datasets efficiently for input to neuromorphic processors. The ability to efficiently encode audio data to spikes will enable a wider variety of spiking audio datasets to be available and can also enable algorithmic development of real-time neuromorphic automatic speech recognition systems. Therefore, we developed speech2spikes, a simple and efficient audio processing pipeline that encodes recorded audio into spikes and is suitable for real-time operation with low-power neuromorphic processors. To demonstrate the efficacy of our method for audio to spike encoding we show that a small feed-forward spiking neural network trained on data generated with the pipeline achieves accuracy on the Google Speech Commands recognition task, exceeding the state-of-the art set by Spiking Speech Commands, a prior spiking encoding of the Google Speech Commands dataset, by over 10%. We also demonstrate a proof-of-concept real-time neuromorphic automatic speech recognition system using audio encoded with speech2spikes streamed to an Intel Loihi neuromorphic research processor.

19 October 2020 in Frontiers in Neurorobotics
Verified
Journal Article
Open access
PDF file
KCKexin Chen
Kexin Chen
THTiffany Hwu
Tiffany Hwu
HKHirak J. Kashyap
Hirak J. Kashyap
JKJeffrey L. Krichmar
Jeffrey L. Krichmar
KSKenneth Stewart
Kenneth Stewart
XZXinyun Zou
Xinyun Zou
Understanding why deep neural networks and machine learning algorithms act as they do is a difficult endeavor. Neuroscientists are faced with similar problems. One way biologists address this issue is by closely observing behavior while recording neurons or manipulating brain circuits. This has been called neuroethology. In a similar way, neurorobotics can be used to explain how neural network activity leads to behavior. In real world settings, neurorobots have been shown to perform behaviors analogous to animals. Moreover, a neuroroboticist has total control over the network, and by analyzing different neural groups or studying the effect of network perturbations (e.g., simulated lesions), they may be able to explain how the robot's behavior arises from artificial brain activity. In this paper, we review neurorobot experiments by focusing on how the robot's behavior leads to a qualitative and quantitative explanation of neural activity, and vice versa, that is, how neural activity leads to behavior. We suggest that using neurorobots as a form of computational neuroethology can be a powerful methodology for understanding neuroscience, as well as for artificial intelligence and machine learning.

31 March 2023 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
TZTianqi Zhang
Tianqi Zhang
JMJusting Morris
Justing Morris
KSKenneth Stewart
Kenneth Stewart
HLHin Wai Lui
Hin Wai Lui
BKBehnam Khaleghi
Behnam Khaleghi
TRTajana Rosing
Tajana Rosing
Today’s machine learning (ML) systems, running workloads, such as deep neural networks, which require billions of parameters and many hours to train a model, consume a significant amount of energy. Due to the complexity of computation and topology, even the quantized models are hard to deploy on edge devices under energy constraints. To combat this, researchers have been focusing on new emerging neuromorphic computing models. Two of those models are hyperdimensional computing (HDC) and spiking neural networks (SNNs), both with their own benefits. HDC has various desirable properties that other ML algorithms lack, such as robustness to noise, simple operations, and high parallelism. SNNs are able to process event-based signal data in an efficient manner. This work develops $\mathsf {HyperSpike}$ , which utilizes a single, randomly initialized, and untrained SNN layer as a feature extractor connected to a trained HDC classifier. HDC is used to enable more efficient classification as well as provide robustness to errors. We experimentally show that $\mathsf {HyperSpike}$ is on average $31.5\times $ more robust to errors than traditional SNNs. On Intel’s Loihi (Davies et al., 2018), $\mathsf {HyperSpike}$ is $10\times $ faster and $2.6\times $ more energy efficient over traditional SNN networks. We further develop $\mathsf {HyperSpikeASIC}$ , a customized accelerator for $\mathsf {HyperSpike}$ . By decoupling the neuron and synapses, $\mathsf {HyperSpikeASIC}$ skips the inactive neurons and limits the neuron state updating to once per time step at most. $\mathsf {HyperSpikeASIC}$ is $601\times $ faster and $3467\times $ more energy efficient than $\mathsf {HyperSpike}$ running on Intel’s Loihi for SNN acceleration, and $12.2\times $ faster and $211\times $ more energy efficient than the state-of-the-art SNN ASIC implementation (Wang et al., 2022).