David Widemann
YOU?
Author Swipe
View article: Photonic Accelerators for Image Segmentation in Autonomous Driving and Defect Detection
Photonic Accelerators for Image Segmentation in Autonomous Driving and Defect Detection Open
Photonic computing promises faster and more energy-efficient deep neural network (DNN) inference than traditional digital hardware. Advances in photonic computing can have profound impacts on applications such as autonomous driving and def…
View article: Latent Space Simulation for Carbon Capture Design Optimization
Latent Space Simulation for Carbon Capture Design Optimization Open
The CO2 capture efficiency in solvent-based carbon capture systems (CCSs) critically depends on the gas-solvent interfacial area (IA), making maximization of IA a foundational challenge in CCS design. While the IA associated with a particu…
View article: Adaptive Block Floating-Point for Analog Deep Learning Hardware
Adaptive Block Floating-Point for Analog Deep Learning Hardware Open
Analog mixed-signal (AMS) devices promise faster, more energy-efficient deep neural network (DNN) inference than their digital counterparts. However, recent studies show that DNNs on AMS devices with fixed-point numbers can incur an accura…
View article: Latent Space Simulation for Carbon Capture Design Optimization
Latent Space Simulation for Carbon Capture Design Optimization Open
The CO2 capture efficiency in solvent-based carbon capture systems (CCSs) critically depends on the gas-solvent interfacial area (IA), making maximization of IA a foundational challenge in CCS design. While the IA associated with a particu…
View article: A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder
A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder Open
View article: MeshGraphNets
MeshGraphNets Open
A PyTorch implementation of "Learning Mesh-based Simulation with Graph Networks" (Pfaff et al., 2021) to build and train graph neural network surrogates for CFD simulation.
View article: Improving five-year survival prediction via multitask learning across HPV-related cancers
Improving five-year survival prediction via multitask learning across HPV-related cancers Open
Oncology is a highly siloed field of research in which sub-disciplinary specialization has limited the amount of information shared between researchers of distinct cancer types. This can be attributed to legitimate differences in the physi…
View article: Efficient nonlinear manifold reduced order model
Efficient nonlinear manifold reduced order model Open
Traditional linear subspace reduced order models (LS-ROMs) are able to accelerate physical simulations, in which the intrinsic solution space falls into a subspace with a small dimension, i.e., the solution space has a small Kolmogorov n-w…
View article: A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder
A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder Open
Traditional linear subspace reduced order models (LS-ROMs) are able to accelerate physical simulations, in which the intrinsic solution space falls into a subspace with a small dimension, i.e., the solution space has a small Kolmogorov n-w…
View article: Multitask Recommender Systems for Cancer Drug Response
Multitask Recommender Systems for Cancer Drug Response Open
The problem we are currently trying to address is that there are many types of cancer drugs and many types of cancers and there is not always experimental data for a specific cancer type and cancer drug interaction. While there is a large …
View article: Bayesian Multitask with Structure Learning
Bayesian Multitask with Structure Learning Open
Software including a Bayesian multitask learning model. The pipeline includes functionality to process data, train, evaluate models, and generate reports.
View article: Multi-Task Learning Software Suite
Multi-Task Learning Software Suite Open
Software suite with multitask learning models. The pipeline includes functionality to process data, train, evaluate models, and generate reports.
View article: Bayesian multitask learning regression for heterogeneous patient cohorts
Bayesian multitask learning regression for heterogeneous patient cohorts Open
View article: Modular Spiking Neural Circuits for Mapping Long Short-Term Memory on a Neurosynaptic Processor
Modular Spiking Neural Circuits for Mapping Long Short-Term Memory on a Neurosynaptic Processor Open
Due to the distributed and asynchronous nature of neural computation through low-energy spikes, brain-inspired hardware systems offer high energy efficiency and massive parallelism. One such platform is the IBM TrueNorth neurosynaptic syst…
View article: TrueNorth Ecosystem for Brain-Inspired Computing: Scalable Systems, Software, and Applications
TrueNorth Ecosystem for Brain-Inspired Computing: Scalable Systems, Software, and Applications Open
View article: ROPE: Recoverable Order-Preserving Embedding of Natural Language
ROPE: Recoverable Order-Preserving Embedding of Natural Language Open
We present a novel Recoverable Order-Preserving Embedding (ROPE) of natural language. ROPE maps natural language passages from sparse concatenated one-hot representations to distributed vector representations of predetermined fixed length.…
View article: Language Model Expansion Using Distributed Representations of Words
Language Model Expansion Using Distributed Representations of Words Open