Search Results for author: Alexander New

Found 11 papers, 3 papers with code

Tunable Complexity Benchmarks for Evaluating Physics-Informed Neural Networks on Coupled Ordinary Differential Equations

no code implementations14 Oct 2022 Alexander New, Benjamin Eng, Andrea C. Timm, Andrew S. Gearhart

In this work, we assess the ability of physics-informed neural networks (PINNs) to solve increasingly-complex coupled ordinary differential equations (ODEs).

Continual learning benefits from multiple sleep mechanisms: NREM, REM, and Synaptic Downscaling

no code implementations9 Sep 2022 Brian S. Robinson, Clare W. Lau, Alexander New, Shane M. Nichols, Erik C. Johnson, Michael Wolmetz, William G. Coon

While some catastrophic forgetting persisted over the course of network training, higher levels of synaptic downscaling lead to better retention of early tasks and further facilitated the recovery of early task accuracy during subsequent training.

Continual Learning Image Classification

Neural Basis Functions for Accelerating Solutions to High Mach Euler Equations

no code implementations2 Aug 2022 David Witman, Alexander New, Hicham Alkendry, Honest Mrema

We propose an approach to solving partial differential equations (PDEs) using a set of neural networks which we call Neural Basis Functions (NBF).

Operator learning Vocal Bursts Intensity Prediction

Curvature-informed multi-task learning for graph networks

no code implementations2 Aug 2022 Alexander New, Michael J. Pekala, Nam Q. Le, Janna Domenico, Christine D. Piatko, Christopher D. Stiles

Properties of interest for crystals and molecules, such as band gap, elasticity, and solubility, are generally related to each other: they are governed by the same underlying laws of physics.

Band Gap Multi-Task Learning

Latent Properties of Lifelong Learning Systems

no code implementations28 Jul 2022 Corban Rivera, Chace Ashcraft, Alexander New, James Schmidt, Gautam Vallabha

Creating artificial intelligence (AI) systems capable of demonstrating lifelong learning is a fundamental challenge, and many approaches and metrics have been proposed to analyze algorithmic properties.

reinforcement-learning Reinforcement Learning (RL)

L2Explorer: A Lifelong Reinforcement Learning Assessment Environment

1 code implementation14 Mar 2022 Erik C. Johnson, Eric Q. Nguyen, Blake Schreurs, Chigozie S. Ewulum, Chace Ashcraft, Neil M. Fendley, Megan M. Baker, Alexander New, Gautam K. Vallabha

Despite groundbreaking progress in reinforcement learning for robotics, gameplay, and other complex domains, major challenges remain in applying reinforcement learning to the evolving, open-world problems often found in critical application spaces.

Continual Learning reinforcement-learning +2

Lifelong Learning Metrics

1 code implementation20 Jan 2022 Alexander New, Megan Baker, Eric Nguyen, Gautam Vallabha

The DARPA Lifelong Learning Machines (L2M) program seeks to yield advances in artificial intelligence (AI) systems so that they are capable of learning (and improving) continuously, leveraging data on one task to improve performance on another, and doing so in a computationally sustainable way.

Autonomous Driving

Semantically-aware population health risk analyses

no code implementations27 Nov 2018 Alexander New, Sabbir M. Rashid, John S. Erickson, Deborah L. McGuinness, Kristin P. Bennett

One primary task of population health analysis is the identification of risk factors that, for some subpopulation, have a significant association with some health condition.

BIG-bench Machine Learning

Cadre Modeling: Simultaneously Discovering Subpopulations and Predictive Models

1 code implementation7 Feb 2018 Alexander New, Curt Breneman, Kristin P. Bennett

In a materials-by-design case study, our model provides state-of-the-art prediction of polymer glass transition temperature.

feature selection regression

Cannot find the paper you are looking for? You can Submit a new open access paper.