Search Publications

Displaying 26 - 44 of 44

Artificial Intelligence Driven Experiments at User Facilities

December 13, 2023

Author(s)

Bruce D. Ravel, Phillip Michael Maffettone, Daniel Allan, Andi Barbour, Thomas Caswell, Dmitri Gavrilov, Marcus Hanwell, Thomas Morris, Daniel Olds, Maksim Rakitin, Stuart Campbell

n/a

Extending Explainable Boosting Machines to Scientific Image Data

November 30, 2023

Author(s)

Daniel Schug, Sai Yerramreddy, Rich Caruana, Craig Greenberg, Justyna Zwolak

As the deployment of computer vision technology becomes increasingly common in science, the need for explanations of the system and its output has become a focus of great concern. Driven by the pressing need for interpretable models in science, we propose

META LEARNING WITH LANGUAGE MODELS: CHALLENGES AND OPPORTUNITIES IN THE CLASSIFICATION OF IMBALANCED TEXT

October 23, 2023

Author(s)

Apostol Vassilev, Honglan Jin, Munawar Hasan

Detecting out of policy speech (OOPS) content is important but difficult. While machine learning is a powerful tool to tackle this challenging task, it is hard to break the performance ceiling due to factors like quantity and quality limitations on

Neural networks three ways: unlocking novel computing schemes using magnetic tunnel junction stochasticity

September 28, 2023

Author(s)

Matthew Daniels, William Borders, Nitin Prasad, Advait Madhavan, Sidra Gibeault, Temitayo Adeyeye, Liam Pocher, Lei Wan, Michael Tran, Jordan Katine, Daniel Lathrop, Brian Hoskins, Tiffany Santos, Patrick Braganca, Mark Stiles, Jabez J. McClelland

Due to their interesting physical properties, myriad operational regimes, small size, and industrial fabrication maturity, magnetic tunnel junctions are uniquely suited for unlocking novel computing schemes for in-hardware neuromorphic computing. In this

Using Deep Reinforcement Learning to Automate Network Configurations for Internet of Vehicles

September 4, 2023

Author(s)

Xing Liu, Cheng Qian, Wei Yu, David W. Griffith, Avi Gopstein, Nada T. Golmie

In this paper, we address the issue of automating network configurations for dynamic network environments such as the Internet of Vehicles (IoV). Configuring network settings in IoV environments has proven to be a difficult task due to their dynamic and

Retrieval of Raman signals from broadband CARS intensity measurements using an autoencoder trained with a prior excitation function

July 31, 2023

Author(s)

ryan muddiman, Kevin O' Dwyer, Charles Camp, Bryan Hennelly

Broadband coherent anti-Stokes Raman scattering (BCARS) is capable of producing high-quality Raman spectra spanning broad bandwidths, 400–4000 cm−1, with millisecond acquisition times. Raw BCARS spectra, however, are a coherent combination of vibrationally

Towards Real-Time Heart Health Monitoring in Firefighting Using Convolutional Neural Networks

June 28, 2023

Author(s)

Jiajia Li, Christopher U. Brown, Dillon Dzikowicz, Mary Carey, Wai Cheong Tam, Michael Xuelin Huang

A machine learning-based heart health monitoring model, named H2M, was developed. 24-hour electrocardiogram (ECG) data from 112 professional firefighters was used to train the proposed model. The model used carefully designed multi-layer convolution neural

Explainable Machine Learning Based Flashover Prediction Model Using Dimension-Wise Class Activation Map

June 26, 2023

Author(s)

Linhao Fan, Wai Cheong Tam, Qi Tong, Eugene Yujun Fu, Tianshui Liang

This paper presents the development of an explainable machine learning based flashover prediction model, named xFlashNet. Synthetic temperature data from more than 17 000 fire cases are used for model development. The effect of missing data due from heat

Multiplexed gradient descent: Fast online training of modern datasets on hardware neural networks without backpropagation

June 26, 2023

Author(s)

Adam McCaughan, Bakhrom Oripov, Natesh Ganesh, Sae Woo Nam, Andrew Dienstfrey, Sonia Buckley

We show that model-free perturbative methods can be used to efficiently train modern neural network architectures in a way that can be directly applied to emerging neuromorphic hardware. These methods were investigated for training VLSI neural networks

Cluster Association for 3D Environment Based on 60 GHz Indoor Channel Measurements

May 31, 2023

Author(s)

Raied Caromi, Jian Wang, Anuraag Bodi, Camillo Gentile

In this paper, we present a ray tracing (RT) assisted multipath cluster association method. This work is based on an indoor channel measurement at 60 GHz, where a light detection and ranging (LiDAR) sensor was co-located with channel sounder and time

AIomics: Exploring More of the Proteome Using Mass Spectral Libraries Extended by Artificial Intelligence

May 26, 2023

Author(s)

Lewis Geer, Joel Lapin, Douglas Slotta, Tytus Mak, Stephen E. Stein

The unbounded permutations of biological molecules, including proteins and their constituent peptides, present a dilemma in identifying the components of complex biosamples. Sequence search algorithms used to identify peptide spectra can be expanded to

Addressing misclassification costs in machine learning through asymmetric loss functions

April 27, 2023

Author(s)

Bryan Barnes, Mark-Alexander Henn

Background: Patterning defect metrology requires data interpretation with classification, each well-suited to machine learning (ML). Defect classification however has notable misclassification costs; mislabeling a defect as nominal has greater impact than

Federated Learning with Server Learning for Non-IID Data

March 24, 2023

Author(s)

Van Sy Mai, Richard La, Tao Zhang, Yu Xuan Huang, Abdella Battou

Self-driving Multimodal Studies at User Facilities

January 22, 2023

Author(s)

Bruce D. Ravel, Phillip Michael Maffettone, Daniel Allan, Stuart Campbell, Matthew Carbone, Brian DeCost, Howie Joress, Dmitri Gavrilov, Marcus Hanwell, Joshua Lynch, Stuart Wilkins, Jakub Wlodek, Daniel Olds

Multimodal characterization is commonly required for understanding materials. User facilities possess the infrastructure to perform these measurements, albeit in serial over days to months. In this paper, we describe a unified multimodal measurement of a

Device Modeling Bias in ReRAM-Based Neural Network Simulations

January 20, 2023

Author(s)

Imtiaz Hossen, Matthew Daniels, Martin Lueker-Boden, Andrew Dienstfrey, Gina Adam, Osama Yousuf

The study of resistive-RAM (ReRAM) devices for energy efficient machine learning accelerators requires fast and robust simulation frameworks that incorporate realistic models of the device population. Jump table modeling has emerged as a phenomenological

Advances and Trends in Machine Learning and Artificial Intelligence: A Polymer Perspective

January 18, 2023

Author(s)

Tyler Martin, Debra Audus

Here we highlight recent advancements in informatics, machine learning, and artificial intelligence as applied to the fields of polymer physics and chemistry. In particular, we focus on the unique challenges presented by polymers and how the field is

Efficient Parameter Exploration of Simulation Studies

November 18, 2022

Author(s)

Megan Olsen, M S Raunak

Simulation is a useful and effective way to analyze and study complex, real-world systems. It allows researchers, practitioners, and decision makers to make sense of the inner working of a system that involves many factors often resulting in some sort of

Advancing Discovery with Artificial Intelligence and Machine Learning at NSLS-II

September 16, 2022

Author(s)

Bruce D. Ravel

With the National Synchrotron Light Source II (NSLS-II) coming online in 2015 as the brightest source in the world, the imminent upgrades at the Advanced Photon Source, Advanced Light Source, and Linear Coherent Light Source, and advances in detector

Finding the Signal from the Smoke: A Real-Time, Unattended Fire Prevention System Using 3D Convolutional Neural Networks

August 31, 2022

Author(s)

Michael Ngai, Eugene Yujun Fu, Wai Cheong Tam, Grace Ngai, Amber Yang

Cooking fires are dangerous. Every year, they are responsible for taking away more than 500 lives in the U.S. alone. Existing approaches using sensors usually require expensive retrofitting and are not feasible in real-life situations. This research